Faculty Biographies

William Y. Arms
Professor
wya@cs.cornell.edu
D.Phil. University of Sussex U.K., 1973

My research interests concentrate on web information systems, digital libraries and electronic publishing. These fields integrate methods from many disciplines, so that the work ranges from technical topics, such as distributed computing and information representation, to the economic and social aspects of change. My book Digital Libraries was published by the MIT Press in winter 2000. This year we received a major grant to integrate many separate projects into the NSF's new digital library for science, mathematics, engineering and technology education. This is likely to be the largest and most heterogeneous digital library yet attempted. Cornell's multidisciplinary team combines computer science, librarian and user interfaces design expertise. One of my principal interests is the change in scientific publication as online materials replace printed journals as the primary means of creating, storing, and distributing research information. I have recently completed a period as chair of the ACM Publications Board, am a member of the MIT Press Management Board, and am a member of a strategic planning committee of the American Physical Society. As part of the NSF-funded Prism project, I am working with the Library of Congress to develop methods for long-term preservation of materials on the Web.


University Activities
Chair, Provost's Advisory Committee on Distance Education. Director, eCornell.
Member of the Faculty Senate, the Faculty Advisory Board on Information Technology, and the University Library Board.

Professional Activities
Publications Board, Association for Computing Machinery.
National Research Council study Issues for Science and Engineering Researchers in the Digital Age.
MIT Press, member of the Management Board and editor of the series on Digital Libraries and Electronic Publishing.

Lectures
The Impact of the Internet on Research Universities. National Science Foundation (April 13, 2001).
The National Science Digital Library Program. Coalition for Networked Information (April 9, 2001).
Quality Control in Scholarly Publishing. What are the Alternatives to Peer Review? Keynote, workshop on the
      Open Archives initiative and peer review journals in Europe, Geneva (March 22 - 24, 2001).
Minerva: The Web Preservation Project. Library of Congress (February 2, 2001).
Strategies for Collecting and Preserving Open Access Materials on the Web. Federal Library and Information
      Center Committee, Washington D.C. (December 7, 2000).
The Web as an Open Access Digital Library. Closing address, 2000 Kyoto International Conference on Digital
      Libraries: Research and Practice, Kyoto, Japan (November 15, 2000).
Open Access to Digital Libraries. Must Research Libraries be Expensive? Keynote address, European
      Conference on Digital Libraries, ECDL2000, Lisbon, Portugal (September 18, 2000).

Publications
"Uniform Resource Names: Handles, PURLs, and Digital Object Identifiers." Communication of the ACM,
      44(5):68 (May, 2001).
"Collecting and Preserving the Web: The Minerva Prototype." RLG DigiNews 5(2) (April2001). With Roger
      Adkins, Cassy Ammen, and Allene Hayes.
      (http://www.rlg.org/preserv/diginews/diginews5-2.html#feature1)
"An Architecture for Reference Linking." Technical Report TR 2000-1820,Computer Science Department,
      Cornell University (October, 2000). With Donna Bergmark, and Carl Lagoze.
"Automated Digital Libraries. How Effectively can Computers be used for the Skilled Tasks of Professional
      Librarianship?" D-Lib Magazine 6(7/8)(July/August, 2000).
      (http://www.dlib.org/dlib/july00/arms/07arms.html )
"Economic Models for Open-access Publishing." IMP(March,2000).(http://www.cisp.org/imp/march_2000
/03_00arms.htm). Digital Libraries. MIT Press, ISBN 0-262-01180-8, 2000.

Graeme Bailey
Professor
bailey@cs.cornell.edu
Ph.D. University of Birmingham U.K., 1977

Originally working in low-dimensional topology and combinatorial group theory, through an odd mixture of circumstances I have become actively involved in research in mathematics and medicine. One of two ongoing research projects in this area is the modeling of lung inflation, together with a research group at the Class One Trauma Center at Upstate Medical Univ., Syracuse, NY. This is in the early stages of a program to extend to various pathologies affecting elasticity and aimed towards effective clinical treatments. The group, now having made some significant advances in answering questions that had remained unsolved for over 30 years, is now in the process of trying to obtain reliable mathematical models. This involves building computer simulations of dynamic packing results under constrained perturbations and deformations. The other project is in understanding deformations of trans membrane proteins used in cell-signaling processes. This is a carefully constrained version of the protein-folding problems that have been exciting the mathematical biology community in recent years; the application of a topological viewpoint in collaborating with molecular pharmacologists and structural biologists has already yielded some intriguing insights.

Honors/Awards
Kenneth A. Goldman '71 Excellence in Teaching Award, 2000.

University Activities
Adjunct Professor; Mathematics.
Member, Fellowship Selection Committees: Rhodes, Marshall, Churchill, and Fulbright.
Member, WCHI-Development and Transition Committee.
Member, Donlon Fellows Development.
Member, Master of Engineering Committee.
Member, Cornell EMS.
Faculty Advisor, Judo Club.

Kenneth P. Birman
Professor
ken@cs.cornell.edu
http://www.cs.cornell.edu/ken/
Ph.D. University of California, Berkeley, 1981

My research is concerned with reliability and security in modern networked environments. This work has three broad themes. Our main focus is on a new system called "Spinglass" (http://www.cs.cornell.edu/Info/Projects/Spinglass ). The idea is to explore a class of reliable multicast protocols that are extremely scalable and provide unusually stable throughput under stress. We believe that stable throughput is a common requirement in demanding critical settings, but few reliable protocols have this property. By scalability, we mean that a system which works with ten computers should also be usable with ten thousand of them. Spinglass involves two subprojects. One, called Astrolabe, is concerned with a new way to represent data in a network. Astrolabe is like a network-wide database in which each computer or component contributes a live tuple. As data change, Astrolabe propagates the updates. The system uses a form of dynamically materialized view to continuously compute summaries of th epicture of the network as a whole. This results in a powerful new tool for distributed monitoring, management, control, and live collaboration. Robbert van Renesse is the leader on this work, and we're collaborating with Al Demers and Johannes Gehrke on aspects related to databases and data mining. The second big part of Spinglass is concerned with reliable multicast. We've developed a scalable multicast protocol that gives probabilistic consistency guarantees, and we are finding ways to apply this in practical settings. Graduate students looking at these questions include Indranil Gupta, who is developing algorithms that make direct use of probabilistic guarantees; Rimon Barr, who is looking at adapting these tools for mobile networks; and Ranveer Chandra and Venugopalen Ramasubramanian, who areinvestigating ad-hoc mobile networking. Kate Jenkins and Ken Hopkinson are looking at applications that arise when using these protocols in real-world settings arising from the restructuring of the electric power grid.
     Our third big activity is joint work with Bob Constable's Nuprl project, and involves the use of formal methods to prove properties of reliable communication protocols, such as those used in Isis, Horus, and Ensemble.
Our project is funded primarily by DARPA, with some additional funding from the Electric Power Research Institute and Microsoft Research. The project is directed by myself, R. van Renesse, and W. Vogels. R. Bhoedjang is visiting as a post-doc for a few years, and developing Intrusion Detection software to make use of Astrolabe.

Honors
Stephen '57 and Marilyn Miles Excellence in Teaching Award, 2000.

University Activities
Committees: Founding Committee for Faculty of Computing and Information Science; University Conflicts of Interest; Chairman of the Responsible Conduct of Research Committee, Engineering College Policy Committee; IP Advisory Council for the Cornell Research Foundation.

Lectures
Next Generation Internet: Unsafe at any Speed?
-. Keynote Speaker: ISDCS '01 (April, 2001).
-. University of Rochester (November, 2000).
-. IBM T.J. Watson Research Center (March, 2000).
-. Keynote: Middleware 2000.

Publications
"Technology Requirements for Virtual Overlay Networks." IEEE Systems, Man and Cybernetics: Special issue
     on Information Assurance (March, 2001).
"Next Generation Internet: Unsafe at Any Speed?" IEEE Computer, Special issue on Infrastructure Protection
     (Fall, 2000).
"Technology Challenges for Virtual Overlay Networks." IEEE Systems, Man, and Cybernetics Information
     Assurance and Security Workshop, West Point, New York (June 6-7, 2000).
"Optimized Group Rekey for Group Communications Systems." Networked and Distributed Systems Security
     2000, San Diego, California. (V). Extended version available as Cornell University, Computer Science TR99-
     1764. With Ohad Rodeh, and Danny Dolev.
"A Dynamic Light-weight Group Service." Journal of Parallel and Distributed Computing 60:1449-1479 (2000).
     With Luis Rodrigues, Katherine Guo, and Paulo Verissimo.
"A Gossip Protocol for Subgroup Multicast." International Workshop on Applied Reliable Group
      Communication (WARGC 2001), Phoenix, AZ (April, 2001). With Kate Jenkins.
"Providing Efficient, Robust Error Recovery through Randomization." International Workshop on Applied
     Reliable Group Communication (WARGC 2001), Phoenix, AZ (April, 2001). With Zhen Xiao.
"Anonymous Gossip: Improving Multicast Reliability in Ad-hoc Networks." International Conference on
     Distributed Computing systems (ICDCS 2001), Phoenix, AZ (April, 2001). With Ranveer Chandra and Vanogupalen Ramasubramanian.
"A Randomized Error Recovery Algorithm for Reliable Multicast." IEEE Infocom 2001 AK (April, 2001). With
     Zhen Xiao.
"Using Epidemic Techniques for Building Ultra-scalable Reliable Communications Systems." Workshop on
     New Visions for Large-scale Networks: Research and Applications, Vienna, VA (March, 2001). With
     Werner Vogels, and Robbert van Renesse.
"Optimizing Buffer Management for Reliable Multicast." Submitted to the 2nd Annual Workshop on Networked
     Group Communication (NGC 2000), Palo Alto, CA (November 8-10, 2000). With Zhen Xiao, and Robbert van
     Renesse.
"Throughput Stability of Reliable Multicast Protocols." ADVIS' 2000, Dokuz Eylul University, Izmir, Turkey
     (October 25-27, 2000). With Oznur Ozkasap.
"A Probabilistically Correct Election Protocol for Large Groups." DISC 2000, Toledo, Spain (October 4-6,
     2000). With Indranil Gupta, and Robbert van Renesse.

Publications-Landmark
"Bimodal Multicast." ACM Transactions on Computer Systems 17(2):41-88 (May, 1999). With M. Hayden, O.
      Ozkasap, Z. Xiao, M. Budiu, and Y. Minsky.
"A Probabilistically Correct Leadership Election Protocol for Large Groups." DISC-2000, Nov 2000, Toledo
      Spain. With I. Gupta and R. van Renesse.

Martin Burtscher
Assistant Professor
Member of the School of Electrical
and Computer Engineering and the Graduate Field of Computer Science
burtscher@csl.cornell.edu
http://www.csl.cornell.edu/~burtscher/
Ph.D. University of Colorado at Boulder, 2000

My research interests are high-performance micro-processor architecture, instruction-level parallelism, and compiler optimizations. In particular, I am exploring hardware- and software-based value prediction, data compression, and latency reduction techniques.
    The constantly widening speed gap between CPUs and memory is becoming more and more of a performance-limiting factor. In fact, current high-end microprocessors already spend a substantial amount of time waiting for memory accesses. To speed up program execution, the CPU needs to process useful instructions while waiting for the memory. One way of providing a processor with useful work is to predict what it will have to do next. Many commodity microprocessors already contain branch predictors to boost their performance, and it is likely that more predictors will be needed to meet the continuing demand for ever-faster CPUs. Designing, evaluating, and improving such predictors is an important focus of my research.
Ongoing projects include locating novel domains that can benefit from prediction, adding compiler support to aid and simplify the prediction hardware, devising means to reduce predictor sizes and power consumption without compromising performance, discovering as-of-yet unobserved patterns to build new predictors, and using value- prediction techniques to enhance branch-prediction accuracy and data-compression rates.

Lectures
Designing a High-performance Load-value Predictor. Hewlett-Packard Company (December, 2000).
The Evolution of a High-performance Load-value Predictor. Lockheed Martin Corporation (November, 2000).
The Design of a High-performance Load Value Predictor. Compaq Computer Corporation (October, 2000).
Hybridizing and Coalescing Load-value Predictors. International Conference on Computer Design (September,
     2000).
Predictability and Exploitability of Load Values. Microsoft Research (June, 2000).

Publications
"Hybridizing and Coalescing Load-value Predictors." International Conference on Computer Design, Austin, TX
     (September, 2000) :81-92. With B. Zorn.
"Exploring Last n Value Prediction." International Conference on Parallel Architectures and Compilation
     Techniques, Paris, France (October, 1999):66-76. With B. Zorn.
"Prediction Outcome History-based Confidence Estimation for Load-value Prediction." Journal of Instruction-
     level Parallelism1 (May 1999). http://www.jilp.org/vol1/  With B. Zorn.

Claire Cardie
Associate Professor
cardie@cs.cornell.edu
http://www.cs.cornell.edu/home/cardie/
Ph.D. University of Massachusetts
Amherst, 1994

My primary research areas are Natural Language Processing (NLP) and Machine Learning (ML) where we focus on developing corpus-based techniques for understanding and extracting information from natural language texts. In particular, my group investigates the use of machine learning techniques as tools for guiding natural language system development and for exploring the mechanisms that underlie language understanding. Our work encompasses three related areas: (1) machine learning of natural language, (2) the use of corpus-based NLP techniques to aid information retrieval (IR) and summarization systems, and (3) the design of user-trainable NLP systems that can efficiently and reliably extract the important information from a document.
In the past year or so we have made progress on both the natural language processing and machine learning aspects of our research. First, we have extended our approach to partial parsing of natural language texts to operate effectively in a weakly supervised learning framework. The original approach, developed with graduate students Scott Mardis and David Pierce, combines corpus-based grammar induction with a very simple pattern-matching algorithm and an optional constituent verification step. In evaluations on a number of large-scale partial parsing tasks involving on-line text, the approach produces partial parsers that are both fast and accurate.
    Unfortunately, however, large amounts of expensive, human-annotated data are required for training. In new work with David Pierce, we investigate the use of weakly supervised learning algorithms for partial parsing that require only a small set of labeled training instances. In particular, we examine the learning behavior of co-training, a weakly supervised learning paradigm in which the redundancy of the learning task is captured by training two classifiers using separate views of the same data. This enables bootstrapping from a small set of labeled training data via a large set of unlabeled data. For noun phrase bracketing, we find that co-training reduces by 36 percent the difference in error between co-trained classifiers and fully supervised classifiers trained on a labeled version of all available data. However, degradation in the quality of the bootstrapped data arises as an obstacle to further improvement. To address this, we propose a moderately supervised variant of co-training in which a human corrects the mistakes made during automatic labeling. Our analysis suggests that corrected co-training and similar moderately supervised methods may help co-training scale to other natural language learning tasks that typically require large amounts of training data.
In a joint project with CoGenTex Inc. and the University of Montreal, we have begun to extend existing corpus-based learning algorithms for information extraction to acquire a broader set of extraction patterns. We have implemented a new learning algorithm, Autoslog-XML, that can extract linguistic entities beyond just noun phrases. Autoslog-XML is a semi-supervised algorithm for locating useful extraction patterns from unrestricted text. The learned extraction patterns have been employed to support domain-specific multi-document summarization in the natural disasters domain. We have also begun to investigate the application of the extraction pattern learning algorithm for Korean texts. Graduate students Vincent Ng and Kiri Wagstaff are part of this joint research effort.
    The techniques described above can, in turn, be used to support a number of practical, end-to-end text-processing tasks in addition to multi-document summarization. For example, we are using the corpus-based partial-parsing techniques as the primary linguistic component in a new system for general-knowledge question answering. The system combines techniques for standard information retrieval, query-dependent text summarization, and shallow syntactic and semantic sentence analysis. In a series of experiments, we examined the role of each statistical and linguistic knowledge source in the question-answering system and find that even very weak linguistic knowledge can offer substantial improvements over purely IR-based techniques for question answering, especially when smoothly integrated with statistical preferences computed by the IR subsystems.
    Finally, in machine learning research with Kiri Wagstaff, we are investigating the use of prior knowledge in the form of user-supplied constraints to improve the performance of clustering algorithms.

University Activities
College Scholar Advisor.
College Scholar Committee Chair.
Independent Major Advisor.
Cognitive Studies Concentration Advisor.
Member: Faculty of Computing and Information Founders; Faculty of Computing and Information Working Group for Information Science; Faculty of Computing and Information Working Group for Crosscutting Education Programs; Independent Major Advisory Board; Provost's Advisory Group of Women in Science and Engineering; Cognitive Studies Steering Committee; Computer Science Department Ph.D. admissions committee; Department of Computer Science space committee; Cognitive Studies Selection Committees for Summer Fellowships, for Continuing Fellowships, and for Incoming Fellowships.

Professional Activities
Secretary: North American Association for Computational Linguistics (2000-2001).
Secretary: SIGNLL, Association for Computational Linguistics special interest group on Natural Language
     Learning (1999-2001).
Editorial Board: Machine Learning (1999-2001), Semantic Web Journal.
Action Editor: Journal of Machine Learning Research (2000-2002).
Program co-chair: Fourth Computational Natural Language Learning Workshop (CoNLL 2000).
Member: DARPA/NSF Question and Answering Roadmap Committee; DARPA/NSF Summarization Roadmap
     Committee; DARPA Translingual Information Detection, Extraction, and Summarization (TIDES) Evaluation Committee.

Program Committees
Eighteenth International Conference on Machine Learning (ICML) (2001).
Seventeenth International Joint Conference on Artificial Intelligence (IJCAI) (2001).
First International Conference on Knowledge Capture (K-CAP) (2001).
39th Annual Meeting of the Association for Computational Linguistics (2001).
Second Meeting of the North American Chapter of the Association for Computational Linguistics (2001).
Fifth Computational Natural Language Learning Workshop (CoNLL) (2001).
Area Chair: The 24th Annual International ACM SIGIR Conference on Research and Development in Information
    Retrieval (2001).
Reviewer: Special issue of Computational Linguistics on anaphora resolution.
Executive Board: SIGDAT, Special Interest Group of ACL for Linguistic Data and Corpus-based approaches to
     NLP.
NSF Review Panel: Knowledge and Cognitive Systems (2000); Human-Computer Interaction (2001).

Lectures
Noun Phrase Co-reference for Information Extraction. Logic and AI Seminar, University of Maryland (April,
     2001).
Machine Learning for Information Extraction from Unrestricted Text. Alan Perlis Symposium, Yale University
     (April, 2001).
Rapidly Portable Translingual Information Extraction and Interactive Multi-document Summarization. DARPA
     TIDES Meeting (February, 2001).
Overview of Cornell University Projects in Natural Language Understanding. America On-line (December, 2000).

Publications
"Limitations of Co-training for Natural Language Learning from Large Datasets." Proceedings of the 2001
     Conference on Empirical Methods in Natural Language Processing (EMNLP-2001), 1-9. Association for
     Computational Linguistics (2001). With David Pierce.
"Multi-document Summarization via Information Extraction." Proceedings of the First International Conference
     on Human Language Technology Research (2001). With Michael White, Tanya Korelsky, Vincent Ng, David
     Pierce, and Kiri Wagstaff.
"Issues, Tasks and Program Structures to Roadmap Research in Question & Answering" (Q&A). DARPA/NSF
     (2000). With J. Burger, V. Chaudhri, R. Gaizauskas, S. Harabagiu, D. Israel, C. Jacquemin, C. Lin, S.
     Maiorano, G. Miller, D. Moldovan, B. Ogden, J. Prager, E. Riloff, A. Singhal, R. Shrihari, T. Strzalkowski,
     E. Voorhees, and R. Weischedel.
"Using Clustering and SuperConcepts within SMART: TREC 6." Information Processing and Management
     36(1):109-131 (2000). With C. Buckley, M. Mitra, and J. Walz.
"A Cognitive Bias Approach to Feature Selection and Weighting for Case-based Learners." Machine Learning
     41:85-116 (2000).
"Examining the Role of Statistical and Linguistic Knowledge Sources in a General-knowledge Question-
     answering System." Proceedings of the Sixth Applied Natural Language Processing Conference (ANLP-
     2000), 180-187. Association for Computational Linguistics / Morgan Kaufmann (2000). With V. Ng, D.
     Pierce, and C. Buckley.
"Clustering with Instance-level Constraints." Proceedings of the Seventeenth International Conference on
     Machine Learning, 1103-1110. Morgan Kaufmann (2000). With Kiri Wagstaff.
Integrating Case-based Learning and Cognitive Biases for Machine Learning of Natural Language." Journal of
     Experimental and Theoretical Artificial Intelligence 11:297-337 (1999).
"The Role of Lexicalization and Pruning for Base Noun Phrase Grammars." Proceedings of the Sixteenth
     National Conference on Artificial Intelligence, 423-430. AAAI Press / MIT Press (1999). With David Pierce.
"Noun Phrase Co-reference as Clustering." Proceedings of the Joint Conference on Empirical Methods in
     Natural Language Processing and Very Large Corpora, 82-89. Association for Computational Linguistics
     (1999). With Kiri Wagstaff.
Error-driven Pruning of Treebank Grammars for Base Noun Phrase Identification." Proceedings of the Annual
     Conference of the Association for Computational Linguistics and COLING-98, 218-224. Association for
     Computational Linguistics (1998). With David Pierce.
"Empirical Methods in Information Extraction." AI Magazine 18(4):65-79 (1997).

[On sabbatic leave 2001-2002]

L. Paul Chew
Senior Research Associate
chew@s.cornell.edu
http://www.cs.cornell.edu/Info/People/chew/chew.html

My primary interest is in geometric algorithms with an emphasis on practical applications. These practical applications have included placement, motion planning, shape comparison, vision, sensing, mesh generation, molecular matching, and protein shape-comparison.
    In recent work on protein shape, Klara Kedem and I have developed a new kind of "consensus shape" for protein families. This is an analog of the consensus string that is sometimes used for multiple alignment of proteins. The idea is based, in part, on our previous work on the Unit-vector Root Mean Square (URMS) distance for protein shapes. The consensus shape is a pseudo-protein with useful properties. It is a pseudo-protein because it fails to have certain characteristics of real proteins. In particular, for the consensus shape, the spacing between successive alpha carbons is variable, with small distances in regions where the members of the protein family exhibit significant variation and large distances (up to the standard spacing of four angstroms) in regions where the family members agree. Despite this nonprotein-like characteristic, the consensus shape does preserve structural information. If all members of a protein family exhibit a geometric relationship between corresponding alpha carbons then that relationship is preserved in the consensus shape. In particular, distances and angles that are consistent across family members are preserved. Thus, the consensus shape provides a compact summary of the significant strucutural information for a family. We are exploring the use of the consensus shape (1) as a tool for improved protein threading for use in protein structure prediction and (2) as a tool for automating the division of proteins into families and subfamilies.
    My work on mesh generation has been motivated by the finite element method for finding approximate solutions to partial differential equations. The first step of this method is to create a mesh, i.e. to divide the given problem region into simple shapes called elements (usually triangles or quadrilaterals in 2D, tetrahedra or hexahedra in 3D). A number of algorithms have been developed to automate this process, but most of them don't guarantee the quality of the resulting mesh (e.g., a triangle may cross a region boundary or there may be some flat triangles, leading to poor error bounds). I developed efficient techniques for producing meshes of guaranteed quality for problems in the plane and for curved surfaces. The triangles produced are close to equilateral in shape; all region boundaries are respected; and the user can control the element density, producing small elements in "interesting" regions and large elements elsewhere.
    I extended this work to produce tetrahedral meshes for three-dimensional problems. The major difficulty here is to avoid producing "slivers": tetrahedra with nicely shaped faces but with near-zero volume. I showed that slivers can be avoided by choosing each new mesh point randomly within a small specified volume. The randomness helps; a good mesh point-one that does not form any slivers-can be found in constant expected time.
    This work is being used in a large, multi-disciplinary project: developing adaptive software for field-driven simulations. In particular, we focus on computational fracture mechanics and reactive, multiphase fluid flows. Our goal is to develop principles for building software systems that can adapt to changing conditions. These conditions include changes in the desired physics (e.g. we may need to change our physics model when we discover that vibration is significant), changes in the desired algorithms (e.g. we may change our solution technique depending on how quickly we are converging toward a solution), and changes in the computing environment (e.g. additional processors may become available or we may lose processors due to processor failure). Other Cornell researchers working on this project are Keshav Pingali, Steve Vavasis, Paul Stodghill, and Tony Ingraffea (Civil Engineering), along with participants at Mississippi State University, Ohio State University, Clark-Atlanta University, and the College of William & Mary.

Professional Activities
Associate Editor: Pattern Recognition: the Journal of the Pattern Recognition Society.
Referee: Journal of the Association for Computing Machinery, SIAM Journal on Scientific Computing.

Publications
"Parallel FEM Simulation of Crack Propagation-challenges, Status, and Perspectives." Proceedings of Irregular
     2000 (2000). With B. Carter, C.S. Chen, G. Heber, A.R. Ingraffea, R. Krause, C. Myers, P.A. Wawrzynek,
     K. Pingali, P. Stodghill, S. Vavasis, N. Chrisochoides, D. Nave, and G.R. Gao.
"Fast Detection of Common Geometric Substructure in Proteins." Journal of Computational Biology 6(3):313-
     325 (1999). With D. Huttenlocher, K. Kedem, and J. Kleinberg.
"Unit-vector RMS (URMS) as a Tool to Analyze Molecular Dynamics Trajectories." Proteins: Structure,
     Function and Genetics 37, 554-564 (1999). With K. Kedem and R. Elber.

Tom Coleman
Professor
Director, Cornell Theory Center
Director, Financial Industry Solutions Center (FISC)

Our research is concerned with the design and understanding of practical and efficient numerical algorithms for continuous optimization problems. Our primary emphasis is the development of algorithms for large-scale optimization, especially as applied to the area of computational finance.
    With colleagues Yuying Li, Peter Mansfield, Arun Verma, and Shirish Chinchalkar, we are developing a variety of tools and methods for computational finance in the areas of portfolio management and options pricing (and hedging). Several Ph.D. students in the Center for Applied Mathematics are also involved in this work: Cristina Patron, Yohan Kim, and Changhong He. Yohan Kim completed his Ph.D. dissertation in May, 2001: Estimation of Smooth Volatility Functions in Option Pricing Models.

University Activities
Director: Cornell Theory Center.
Director: Financial Industry Solutions Center (55 Broad Street, NYC).
Member: Engineering Dean Search Committee; Cornell Task Force on Genomics; Program Committee for the Center for Applied Mathematics.

Professional Activities
Chair: SIAM Activity Group on Optimization.
Co-organizer: The 11th Annual Derivatives Securities Conference, New York (April 27-28).
Organizer: FISC Spring 2000 Workshop Series.
Program Committee: Automatic Differentiation 2000, INRIA, France.
Member: Advisory Board, Brookhaven Center for Data Intensive Computing; Scientific Program Committee.
Member: Organizing committee for the thematic year at the Fields Institute: Numerical and Computational
     Challenges in Science and Engineering (2001-2002).
Organizer: Proposal for an IMA Special Year on Optimization, Minneapolis, MN (2002-2003); Seventh SIAM
     Conference on Optimization, Toronto, Ontario (2002).
Editorial Board: Applied Mathematics Letters; SIAM Journal of Scientific Computing; Computational
     Optimization and Applications, Comm. on Applied Non-linear Analysis, Mathematical Modeling and
     Scientific Computing.
Referee/Reviewer: Mathematical Programming; Computational Optimization and Applications; SIAM Journal of
     Optimization; SIAM Journal of Scientific Computing; Department of Energy, NSF.

Lectures
Dynamic Hedging and a Deterministic Local Implied Volatility Function. Eleventh Annual Derivatives
     Conference, New York (April 28, 2001).
RISK 2001 Europe, Paris France (April 10-11, 2001).
Introduction to Computational Finance in Matlab: A 1-day workshop at FISC-Japan, Tokyo (March 22, 2001).
A Newton Method for Option Valuation. U. Singapore, Singapore (December 12, 2000).
-. City University of Hong Kong, Hong Kong (December 22, 2000).

Publications
"ADMIT?1: Automatic Differentiation and MATLAB Interface Toolbox." ACM Transactions on Mathematical
     Software 22:150-175 (2000). With Arun Verma.
"Efficiency Improvements for Pricing American Options with a Stochastic Mesh: Parallel Implementation."
     Financial Engineering News, 1-2 (December, 2000). With Thanos Avranidis, Yuriy Zinchenko, and Arun
     Verma.

[On sabbatic leave 2001-2002]

Bob Constable
Professor
Dean for Computing and Information Science
rc@cs.cornell.edu
http://www.cs.cornell.edu/home/rc/
Ph.D. University of Wisconsin-Madison, 1968

We have been building a system that we call a logical programming environment (LPE). It provides substantial automation in the design, coding, verification, and evolution of large software systems. Generally an LPE will integrate programming languages and logics. In our case we integrate the ML programming language and a programming logic based on type theory. Reasoning about ML programs is founded on type theoretic semantics for ML. The LPE also integrates a compiler and a theorem prover. We use the latest version of Nuprl as the prover.
    Over the past year we have deployed our prototype LPE to support the Ensemble group communication system that Ken Birman, Robbert van Renesse, and their students have built. The LPE provides automatic code transformations that improve performance in a way that is guaranteed not to introduce errors. The LPE also supports the design and coding of new adaptive protocols that are part of the Spinglass project.
    The Nuprl 5 system is a major re-implementation of Nuprl 4; its design is based on communicating processes that synchronize around a logical database that we call "the Library." The Library stores definitions, theorems, conjectures, proofs, algorithms, proof tactics, and commentary that is linked to the formal mathematics. We released the first version of Nuprl 5 last summer, with a debut at CADE in June involving the whole design and implementation team (Stuart Allen, Rich Eaton, Christoph Kreitz, Lori Lorigo, and me). Nuprl 5 supports multiple proof engines and multiple editors. It also integrates an automatic theorem prover, called the JProver, built by C. Kreitz, Jens Otten, and Stephan Schmitt.
    The Library includes over ten thousand theorems. Many of these are used in system verification, but a large number are from general mathematics. These general theorems are a valuable resource. We are funded by ONR to further develop and explore the concept of a formal digital library of formal constructive mathematics built around these theorems. We will generalize our mechanisms to allow many users to contribute to the Library using other theorem provers such as ACL2, HOL and PVS.
    We are also working on a more experimental LPE called MetaPRL, which started with Jason Hickey's thesis and now involves many people but especially Aleksey Nogin and Alexsey Kopylov. This system is built entirely in OCaml and supports OCaml as its programming language. It is coded for efficiency as well as modularity, and at some tasks it is over two orders of magnitude faster than Nuprl 5. It also supports multiple programming logics. Nuprl 5 and MetaPRL share mathematics libraries.
    The two theorem provers are used in a variety of other projects as well, including the creation of formal courseware by S. Allen, the translation of formal proofs into natural language by Amanda Holland-Minkley, the automatic analysis of the computational complexity of higher-order programs by Ralph Benzinger, and efficient reflection being designed and implemented by Eli Barzilay. We follow the work of Greg Morrisett and his students on new ML compilers and on typed assembly language. We plan to use the LPE to broadly support research on language-based security in the department and at the new Information Assurance Institute, including the work of Dexter Kozen, Andrew Myers, and Fred Schneider.

University Activities
Dean: Computing and Information Science.
Committees: Applied Math Policy; Cognitive Studies Executive Committee; Theory Center Executive Committee.

Professional Activities
Advisory Council: Princeton University Department of Computer Science.
Editor: Journal of Logic and Computation; Formal Methods in System Design; Journal of Symbolic
    Computation.
Director: NATO Summer School, Marktoberdorf, Germany.
General Committee Member: LICS.

Lectures
How Computers Think. Dean's Forum, Faculty of Natural Sciences, Ben-Gurion University (June, 2001).
Formal Complexity Classes: An Approach to Automating Computational Complexity Analysis. Cornell
     University (May, 2001). With Ralph Benzinger.
How Nuprl Reasons. University of Delaware (May, 2001).
Taking a MEGABYTE: Cornell in the Information Age. University of Delaware (May, 2001).
How Nuprl Reasons. Yale University (February, 2001).
Taking a MEGABYTE: Cornell in the Information Age. Cornell University (February, 2001).
Representing the Faculty of Computing and Information. Cornell University (September, 2000).
How Computers Think. Cornell University (September, 2000).
Computer Science: Achievements and Challenges Circa 2000. Cornell University (March, 2000).

Publications
"An Experiment in Formal Design Using Meta-properties. In Proceedings of DARPA Information Survivability
     Conference and Exposition II (DISCEX 2001), IEEE Computer Society Press (June, 2001). With C. Kreitz,
     M. Bickford, and R. van Renesse.
"Protocol Switching: Exploiting Meta-properties." In Proceedings of the International Workshop on Applied
     Reliable Group Communication (WARGC 2001) (April, 2001). IEEE Computer Society Press. With C.
     Kreitz, X. Liu, R. van Renesse, and M. Bickford.
"Computational Complexity and Induction for Partial Computable Functions in Type Theory." In Reflections: A
     Collection of Essays in Honor of Solomon Feferman, Association for Symbolic Logic (2001). With K. Crary.
"Constructively Formalizing Automata." In Proof, Language and Interaction: Essays in Honour of Robin Milner,
     MIT Press, 213-238 (2000). With P.B. Jackson, P. Naumov, and J. Uribe.
"Nuprl's Class Theory and its Applications." In Foundations of Secure Computation, F.L. Bauer and R.
     Steinbruggen, editors. IOS Press, 91-115 (2000).
"Types in Logic, Mathematics and Programming." In Handbook of Proof Theory, S. R. Buss, editor. Elsevier
     Science B. V., 684-785 (1998).
"The Structure of Nuprl's Type Theory." In Logic of Computation, Springer-Verlag, (1997).
Implementing Mathematics with the Nuprl Development System. Prentice-Hall (1986). With S. Allen, et. al.

Alan Demers
Professor
ademers@cs.cornell.edu
Ph.D. Princeton University, 1975

My current research concerns aspects of weakly-consistent data replication in databases and distributed systems. With Ken Birman, Robbert van Renesse, Johannes Gehrke, and others, I am studying randomized "gossip protocols." Such protocols are highly fault-tolerant and, when properly designed, extremely scalable as well. We are studying convergence properties of several flat and hierarchal versions of the basic protocols tailored to specific application requirements.
    My particular focus is approximate evaluation of aggregate queries in such a system. We are studying age distributions of gossiped data in order to prove probabilistic bounds on the quality of aggregate query results. Alternatively, we can use this approach to bound the latency required to probabilistically guarantee a client-specified degree of consistency.
    We are considering classes of "resource location" or "anomaly detection" problems, in which query results are site-specific and depend on distance and subsumption. Such problems can have efficient and highly scalable solutions using gossip partner choice distributions based on the distance between sites.
    Finally, we are studying graph constructions for which flooding or deterministic gossip partner choices can be used, leading to reduced overhead while still retaining most of the desirable properties of randomized gossip.
    The above is related to my previous work on the Clearinghouse and Bayou projects at Xerox PARC. I am also doing work supported by Oracle on asynchronous update-anywhere replication in a more traditional database setting. This involves algorithms for scheduling/reordering update propagation between sites to improve throughput while preserving eventual consistency and bounded inconsistency during propagation.

Recent Papers
"Logarithmic Harary Graphs." ICDCS International Workshop on Applied Reliable Group Communication,
     Phoenix, Arizona (April, 2001). With K. Jenkins.
"Spatial Gossip and Resource Location Protocols." Proceedings of the 33d ACM Symposium on Theory of
     Computing, Crete (July, 2001). With D. Kempe, and J. Kleinberg.

Ron Elber
Professor
ron@cs.cornell.edu
http://www.cs.cornell.edu/ron/old_webpage-ron/index_ron.html
Ph.D. Hebrew University, Jerusalem, 1984

My research is in the field of Computational Molecular Biology. We develop computer algorithms to study sequences, structures, dynamics, and function of proteins and apply these methods to a variety of biological problems. Our techniques are implemented in the systems MOIL and LOOPP, available on the web http://www.tc.cornell.edu/CBIO.
    My current research directions include: mean field approaches for global optimization and structure prediction (Locally Enhanced Sampling). Structures are often determined by an optimization of an energy function. I introduced mean field approaches that modify the target function and make it more accessible to global optimization. We have applied these techniques to determine conformations of short peptides and to refine low-resolution structures of proteins. These approaches are implemented into MOIL.
    We are also working on development of folding potentials using linear programming. An ideal folding potential assigns the lowest energy to the correct three-dimensional structure of a protein. All other structures must have higher energies. The design of folding potentials relies on considerable human intuition and many trials and errors. I developed an automated protocol that "learns" and improves the quality of the current potential energy. We used this protocol to prove that the widely used pairwise interactionmodel cannot recognize exactly correct protein folds. Based on these studies, a novel threading algorithm was designed and implemented in the program LOOPP. In a threading algorithm a sequence is matched with a structure. In a recent article in Science, we published an intriguing application of this program. We suggested an evolutionary link between a gene that controls the size of the tomato fruit and a protein that participates in controlling cell growth and division. Malfunction of this protein causes cancer in humans (joint work with Steve Tanksley's group).
    Another project concerns extending the time scale of simulations. One of the striking observations in dynamics of biological molecules is the extremely large time scale they covered. Initiation by light absorption of biochemical processes is very rapid (femtoseconds), while protein folding is slow (milliseconds to minutes). Current simulation approaches (Molecular Dynamics (MD)) are restricted to nanoseconds (10-9 seconds). I developed a stochastic path integral formulation that provides a numerically stable trajectory for almost any arbitrary time step. We apply the new algorithm to study activation of proteins (the R->T transitions in hemoglobin, microseconds) and to protein folding (folding of C peptide, tens of nanoseconds). The method provides systematic approximations to the dynamics and is more efficient than MD by orders of magnitude. It is available in MOIL.

Professional Activities
Acting head: NIH resource for parallel computing at the Cornell Theory Center.
Committees: Statistical and Computational Genomics Committee; Computational Biology Committee for the collaborative efforts at Cornell, Rockefeller, and Sloan Kettering Institutes; Cornell Life Science Advisory Board;
     Planning Committee for Life Science and Technology Building; Theory Center committee.
National Committees: NIH study sections; NSF study section; Reviewer for the State of Texas; Chair:
     workshop on protein dynamics, Telluride (August, 2001).

Lectures
Protein Recognition by Threading. DIMACS, Rutgers (March, 2001).
Parallel Computations of Trajectories. SIAM (March, 2001).
Long Time Dynamics of Proteins. University of Pennsylvania (February, 2001).
Long Time Dynamics and Protein Recognition by Threading. IBM Watson (February, 2001).
Long Time Dynamics of Biomolecules. Florida State, Computational Biophysics (January, 2001).
Protein Recognition by Threading. CUNY (December, 2000).
Protein Recognition by Threading. University of Maryland (October, 2000).
Long Time Dynamics of Biomolecules. NYU (October, 2000), M3.

Publications
"Protein Recognition by Sequence-to-structure Fitness: Bridging Efficiency and Capacity of Threading Models."
     Submitted to Advances in Chemical Physics, by invitation. With Jaroslaw Meller.
"The Enzymatic Circularization of a Malto-octaose Linear Chain Studied by Stochastic Reaction Path
     Calculations on Cyclodextrin Glycosyltransferase." Proteins, Structure, Function and Genetics 43:327-335 (2001). With Joost C.M. Uitdehaag, Bart A. van der Veen, Lubbert Dijkhuizen, and Bauke W. Dijkstra.
"Cloning, Transgenic Expression and Function of fw2.2: a Quantitative Trait Locus Key to the Evolution of
     Tomato Fruit." Science 289:85-88 (2000). With A. Frary, C. Nesbitt, A. Frary, S. Grandillo, E. van der
     Knaap, B. Cong, J. Liu, J. Meller, K.B. Alpert, and S.D. Tanksley.
"Distance Dependent Pair Potential for Protein Folding: Results from Linear Optimization." Proteins, Structure,
     Function and Genetics 41:40-46 (2000). With D. Tobi.
"Probing the Role of the Local Propensity in Peptide Turn Formation." International Journal of Quantum
     Chemistry 80:1125-1128 (2000). With D. Mohanty, and D. Thirumalai.

Geri Gay
Professor
FCI, Joint with Communications
gkg1@cornell.edu
http://www.comm.cornell.edu/faculty/gay.html
Ph.D. Cornell University, 1985

Iam the director of the Human Computer Interaction Group (HCI Group) and a professor of communication at Cornell University. The HCI Group is a research and development group whose members design and research the use of computer-mediated learning environments. My research interests focus on cognitive and social issues for the design and use of interactive communication technologies. Past research has explored navigation issues, knowledge management, mental models and metaphors, knowledge representations, collaborative work and learning, and system design.
     I have received funding for my research and design projects from the National Science Foundation (NSF), the National Endowment for the Humanities (NEH), the Mellon Foundation, Intel, Microsoft, IBM, Getty, and several private donors. I teach courses in interactive multimedia design and research, computer-mediated communication, human-computer interaction, and the social design of communication systems.

Honors/Awards
NYS Chancellor's Award for Excellence 2001.
Innovative Teaching Award 2000.

Current Projects
NASA and AT&T Advanced Technology for Learning projects.
Intel Museum Context Aware Computing Project.
Intel and NSF studies on the use of wireless computing, covered in Chronicle of Higher Education, USA Today,
      Newsweek, The New York Times, Globe and Mail, and NPR.

Lectures
American Educational Research Association, International Communication Association; ACM Multimedia; Japanese Private University Association.

Publications
Articles in Journal of Computer-Mediated Communication; Journal of Research on Computing in Education; Journal of Educational Technology; Journal of Educational Computing Research; Journal of Educational Psychology; International Journal of HCI; ACM Digital Libraries; Journal of Information Technologies.

Johannes Gehrke
Assistant Professor
johannes@cs.cornell.edu
http://www.cs.cornell.edu/johannes/
Ph.D. University of Wisconsin-Madison, 1999

My primary research interest is in the development of new data mining and database technology. My group is currently involved in three projects: The Himalaya Data Mining Project, the Cougar Sensor Database System, and the Amazon Stream Processing Project.
     In the Himalaya Data Mining Project we develop new data mining functionality, and we work on techniques to make the resulting data mining models more understandable to the user. As an example, consider classification trees, a data mining model that is supported in nearly all commercial data mining suites. In recent research we have shown that a large class of classification tree construction algorithms is biased (including most algorithms used in commercial tools), thus, users could draw incorrect conclusions from the resulting "incorrect" classification tree. Our methods can provably eliminate this bias from any existing split selection method. Other recent results include the fastest published algorithm for mining long market baskets, and new methods for mining long sequential patterns.
     The widespread deployment of sensors and mobile devices is transforming our physical environment into a computing platform. There is now computing power on every device, and emerging networking techniques ensure that devices are interconnected and accessible from local or wide-area networks. This is a distributed database system of unprecedented scale. In the Cougar Sensor DatabaseSystem, we develop database technology for tasking, mining, and monitoring such a large number of distributed data sources. We have implemented the first generation of the Cougar Device Database System, where we leverage the processing power on the devices to push query processing directly to the data sources. Different query processing strategies allow us to balance resource usage, accuracy, and speed of query answers. Our current research focuses on distributed and fault tolerant query processing and meta-data management.
In many applications, for example in intrusion detection, sensor networks, and network management, data arrive in streams, and the large volume of such high-speed data streams makes storage and offline processing of the data infeasible. In the Amazon Stream Processing Project, we are developing query processing techniques for long running queries over infinite data streams. The main difficulty here is the new model of computation: Instead of being able to re-read data many times and to perform expensive offline computation on a static dataset, we need to compute query answers and maintain summary statistics in an online fashion. Our recent results include computation of correlated aggregates and quantiles over data streams.

Honors/Awards
IBM Faculty Development Award (2000, 2001).
James and Mary Tien Excellence in Teaching Award (2001).

University Activities
Member: Space Committee, Department of Computer Science; Faculty Search Committee, School of
      Operations Research and Industrial Engineering; faculty Search Committee, The Computational and
      Statistical Genomics Trust.

Profesional Activities
Program Committees:
Twenty-sixth International Conference on Very Large Databases (VLDB), Cairo, Egypt (September, 2000).
      Demonstrations Committee.
Seventeenth IEEE International Conference on Data Engineering (ICDE 2001), Heidelberg, Germany (April,
      2001).
Fourth International Workshop on Parallel and Distributed Data Mining. Workshop held in conjunction with the
      15th International Parallel and Distributed Processing Symposium, San Francisco, CA (April, 2001).
Twentieth ACM SIGMOD Conference (SIGMOD 2001), Santa Barbara, CA (May, 2001).
ACM SIGMOD Workshop on Research Issues in Data Mining and Knowledge Discovery (DMKD 2001) held in
      cooperation with SIGMOD 2001. Santa Barbara, CA (May, 2001). Workshop Co-chair.
Eighteenth International Conference on Machine Learning (ICML 2001), Williams College, MA (June, 2001). Sixth ACM SIGKDD Conference (KDD 2001), San Jose, CA (August, 2001).
Twelfth International Conference on Software Engineering and Knowledge Engineering, Chicago, IL (July, 2000).
Sixth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2000). Boston,
      MA (August, 2000).
Editorial Boards:
Knowledge and Information Systems.
Journal of Database Management.

Lectures
An Introduction to Data Mining. Air Force Research Laboratory, Rome, NY (September 12, 2000).
An Overview of Modern Data Mining Technology. Workshop at the Financial Industry Solutions Center (FISC)
      New York (November 8, 2000).
The Infrastructure of Electronic Commerce. Lectures on Database Technology and Data Mining. Johnson
      Graduate School of Management, Cornell University (January, 2001).
Honest Classification Trees. IBM Watson Research Center, Yorktown, NY (March, 2001).
Panel Manager for "Storage-A Crowded Place?" Panel at the 2001 Leadership in the Technology Marketplace
      Symposium, Cornell Johnson Graduate School of Management, Ithaca (April, 2001).
Querying the Physical World. DARPA Sensor Information Technology PI Meeting. St. Petersburg, FL (April,
      2001).
Mining Very Large Databases. Invited talk at the 33rd Symposium on the Interface of Computing Science and
      Statistics, Costa Mesa, CA (June, 2001).

Publications
"RAINFOREST - A Framework for Fast Decision Tree Construction of Large Datasets. In Data Mining and
      Knowledge Discovery 4(2/3):27-162 (July, 2000). With Raghu Ramakrishnan, and Venkatesh Ganti.
"Querying the Physical World." In IEEE Personal Communications, special issue on Smart Spaces and
      Environments (October, 2000). With Philippe Bonnet, and Praveen Seshadri.
"Towards Sensor Database Systems." In Proceedings of the Second International Conference on Mobile Data
      Management, Hong Kong, China (January, 2001). With Philippe Bonnet, and Praveen Seshadri.
"DEMON: Mining and Monitoring Evolving Data." In IEEE Transactions on Knowledge and Data Engineering      13(1):50-63 (January/February 2001). With Venkatesh Ganti, and Raghu Ramakrishnan.
"MAFIA: A Maximal Frequent Itemset Algorithm for Transactional Databases." In Proceedings of the 17th
      International Conference on Data Engineering, Heidelberg, Germany (April, 2001). With Doug Burdick, and
      Manuel Calimlim.
"On Computing Correlated Aggregates Over Continual Data Streams." In Proceedings of the 2001 ACM
      SIGMOD International Conference on Management of Data, Santa Barbara, CA (May, 2001). With Flip
      Korn, and Divesh Srivastava.
"Query Optimization In Compressed Database Systems." In Proceedings of the 2001 ACM SIGMOD
      International Conference on Management of Data, Santa Barbara, CA (May, 2001). With Zhiyuan Chen,
      and Flip Korn.
"Bias Correction in Classification Tree Construction." In Proceedings of the Seventeenth International
      Conference on Machine Learning (ICML 2001), Williams College, MA (June, 2001). With Alin Dobra.

Carla Gomes
Research Associate
Director, Intelligent Information Systems Institute (IISI)
gomes@cs.cornell.edu
http://www.cs.cornell.edu/gomes
Ph.D. University of Edinburgh, 1993

My research interests are centered around the integration of methods from artificial intelligence and operations research for solving hard combinatorial problems. I consider applications in areas ranging from combinatorial design, planning and scheduling, reasoning, multi-agent systems, and machine learning. Recently, I have focused on randomized search techniques. In this work, I study so-called heavy-tailed distributions that characterize complete randomized search methods. A promising way of exploiting heavy-tailed behavior is by combining a suite of search methods into a portfolio, running on a distributed compute cluster. It can be shown that such portfolios dramatically reduce the expected overall computational cost, thereby allowing us to solve large, previously unsolved combinatorial problems. Another recent research direction (joint work with groups at the University of Washington and Microsoft Research) involves the use of machine learning techniques and Bayesian models to develop effective adaptive algorithmic strategies given bounded computational resources.
I also established and direct the newly formed Intelligent Information Systems Institute (IISI) at Cornell. The mission of the institute is to foster research in computation and data intensive methods for intelligent decision making systems. See www.cis.cornell.edu/iisi.

Professional Activities
Director, Intelligent Information Systems Institute (IISI).
Guest Editor, Journal of Knowledge Engineering Review, Cambridge Press.
Guest Editor, Artificial Intelligence Journal.
Editorial Board, Journal of Knowledge Engineering Review.
Editorial Board, International Journal AI Tools (IJAIT).
Co-chair, AAAI Symposium on Uncertainty in Computation, Boston, MA (2001).
Co-chair, AAAI Workshop on Leveraging Probability and Uncertainty in Computation, AAAI (2000).
Member, Advisory Committee International Scientists, Ministry of Science and Technology, Portuguese Government, Presidency of European Union (2000).
Member, Program Committee, SAT, Boston, MA (2001).
External Examiner, Ph.D. Thesis of Ramon Bejar, Univ. of Leida, Barcelona, Spain.
Reviewer for 7th Int. Joint Conference on Artificial Intelligence (IJCAI); Journal of Automated Reasoning; Journal
of Artificial Intelligence Research; Constraints: An International Journal; Discrete Applied Mathematics.

Lectures
Structure, Duality, and Randomization: Common Themes in AI and OR.
-. Broad Area Colloquium, Stanford University, Stanford, CA (November, 2000).
-. AI Seminar, SRI, Palo Alto, CA (November, 2000).
-. Research Seminar, NASA/Ames, Mountain View, CA (November, 2000).
-. Colloquium, Univ. of Lisbon, Lisbon, Portugal (March, 2001).
Survey of Information Retrieval and Knowledge Representation, 3 lectures at AFRL/IF, Rome, NY (November,
      2000).
Vision and Directions for the Intelligent Information System Institute, AFRL/IF, Rome, NY (February, 2001).
Impact of Structure on Complexity. MURI /AFOSR meeting on Coop. Control of Distributed Autonomous
      Vehicles in Adversarial Environments, UCLA, Los Angeles, CA (May, 2001).
Structure and Complexity in the Virtual Transportation Company. Meeting on Taskable Agent Software Kit,
      DARPA, Sante Fe, NM (April, 2001).
Controlling Computational Cost. Meeting on Autonomous Negotiation Teams, DARPA, Lake Tahoe, CA (May,
      2001).

Publications
"On the Intersection of AI and OR." Journal of Knowledge Engineering Review 16(1) (2001).
"Algorithm Portfolios." Artificial Intelligence Journal 126(2001). With B. Selman.
"A Bayesian Approach to Tackling Hard Computational Problems." Proc. 17th Conf. on Uncertainty and
      Artificial Intelligence (UAI-2001), Seattle, WA (2001). With E. Horvitz, Y. Ruan, H. Kautz, B. Selman, and      M. Chickering.
"Balance and Filtering in Structured Satisfiable Problems. Proc. 17th Intl. Conf. on Artificial Intelligence (IJCAI-
      2001), Seattle, WA (2001). With H. Kautz, Y. Ruan, D. Achlioptas, B. Selman, and M. Stickel.
"Extending the Reach of SAT with Many-valued Logics." Electronic Notes in Discrete Mathematics 9, Elsevier
      Science Publ. (2001). With R. Bejar, A. Cabiscol, C. Fernandez, and F. Manya.
"An Application of Randomization and Restarts in Proof Planning." Proc. of the 6th European Conference on
      Planning (ECP-01), Toledo, Spain (2001). With A. Meier, and E.Melis.
"Extending the Reach of Proof Planning by Randomization and Restart Techniques." Future Directions in
      Automated Reasoning, IJCAR Workshop, Siena, Italy (2001). With A. Meier and E. Melis.
"Generating Hard Feasible Schedules. Proc. of the 6th European Conference on Planning (ECP-01), Toledo,
      Spain (2001). With J. Argelich, R. Bejar, A. Cabiscol, C. Fernandez, and F. Manya.
"Heavy-tailed Behavior and Randomization in Proof Planning. Model-based Validation of Intelligence, AAAI
      2001 Spring Symposium Series, Stanford, CA (2001). With A. Meier and E. Melis.
"Distribute Constraint Satisfaction in a Wireless Sensor Tracking System. Proc. Workshop on Distributed
      Constraint Reasoning (CONS-2), IJCAI-2001, Seattle (2001). With R. Bejar, B. Krishnamachari, and B.
      Selman.
"Heavy-tailed Phenomena in Satisfiability and Constraint Satisfaction Problems. Journal of Automated
      Reasoning 24(1/2):67-100 (2000). With B. Selman, N. Crato, and Henry Kautz). Results featured in
      Science News (2000).
"Artificial Intelligence and Operations Research: Challenges and Opportunities in Planning and Scheduling and
      Operations Research. Journal of Knowledge Engineering Review 15(1) (2000).

Donald Greenberg
Professor and Member of the FCI,
the Johnson School of Management,
the Department of Architecture, and the
Graduate Field of Computer Science
dpg@graphics.cornell.edu
http://www.graphics.cornell.edu/people/director.html
Ph.D. Cornell University, 1968

Iam one of the pioneers in the emerging field of computer graphics, having served as a leading researcher and teacher in the field since 1965. My research is primarily concerned with physically based image synthesis and with applying graphic techniques to a variety of disciplines. My specialties include color science, parallel processing, and realistic image generation. My application work now focuses on medical imaging, architectural design, perception, digital photography, and real-time photorealistic image generation.
     Consistent with the interdisciplinary nature of the field of computer graphics, I am a member of Cornell's faculty in Johnson Graduate School of Management, the Department of Computer Science, and the Department of Architecture. In recent years I have taught courses in computer graphics, computer-aided architectural design, digital photography, and disruptive technologies.
     I was the founding director of the NSF Science and Technology Center for Computer Graphics and Scientific Visualization, now in its eleventh year. I am the director of the Program of Computer Graphics and former director of the Computer-aided Design Instructional Facility at Cornell.
     I have published more than 200 articles on computer graphics, and many of my students have been highly recognized in the field, including several who have received the SIGGRAPH Achievement Award and others who have received Hollywood Oscars.
     In 1987, I received the ACM Steven Coons Award, the highest honor in the field, for my outstanding creative contributions in computer graphics. I also received the National Computer Graphics Association Academic Award in 1989. In 1997 I received the ASCA Creative Research Award in Architecture. An honorary doctoral degree from New Jersey Institute of Technology was presented to me in 1999.

Professional Activities
Member: National Academy of Engineering.
Fellow: ACM and International Association of Medical and Biological Engineering.

Lectures
Disruptive Histories of Our Future. Cornell JGSM Reunion 2001, Cornell University (June 9, 2001).
Art & Design for Living. Cornell University, Parents Visit (April 20, 2001).
Virtual Environments of the 21st Century. Cornell Association of Professors Emeriti, Cornell University (April
      19, 2001).
Rendering History & Progress. STC Lecture, Program of Computer Graphics, Cornell University (April 17,
      2001).
The Real Challenge for Architecture in a Virtual World. Architecture Department, University of Virginia,
      Charlottesville, VA (April 5, 2001).
Rendering. EAG 2001, Boston, MA (April 3, 2001).
Working Today on Tomorrow's Design Software. SOM Lecture Series, NY (February 15, 2001).
B-schools: A Case History of Our Future. MBA Leadership 2001 Conference, New Orleans, Louisiana (January
      25, 2001).
Distance Learning. Cornell Club of Eastern Florida, Delray Beach, FL (December 6, 2000).
Great Ideas for Computer Science. Lecture, ComS 150, Computer Science, Cornell University (October 18,
      2000).
Technology & Design Practices. Real-time Workshop on Technology and Design Practice, Program of
      Computer Graphics, Cornell University (October 8, 2000).
Tomorrow's Internet and Design Software. Proceedings: Emerging Information Technologies for Facilities, NAS,
      FCC, Washington D.C. (October 20, 2000).
How Do We Prepare Our Students for the Future? Architecture Summer School: Architecture Department,
      Cornell University (July 12, 2000).

Publications
"Spatiotemporal Sensitivity and Visual Attention For Efficient Rendering of Dynamic Environments." ACM
      Transactions on Graphics (2001). With Hector (Yang-Li) Yee, and Sumanta N. Pattanaik.
"Field Trip to Ithaca, N.Y.: Autodesk Development Team Refining Sketching Advantages of Architectural
      Studio." Design Architecture.com (June 4, 2001). With Wendy Talarico.
"Lighting the Way: A Conversation with Don Greenberg of Cornell's Program in Computer Graphics." Cadence
      Web (http://www.cadenceweb.com/features/interviews/greenberg.html), 1-2 (2001).
"Tomorrow's Internet and Design Software." Symposium: Emerging Information Technologies for Facilities,
      NAS, FCC, Washington D.C. (October 20, 2000).
"Once and Future Graphics Pioneer." Architecture Week (18):1-7 (September, 2000). With B. J. Novitski.
"Time-dependent Visual Adaptation for Realistic Image Display." Computer Graphics Proceedings, Annual
      Conference Series, ACM SIGGRAPH, 47-53(July, 2000). With Sumanta N. Pattanaik, Jack Tumblin, and Hector Yee.
"Toward a Psychophysically Based Light Reflection Model for Image Synthesis." Computer Graphics Proceedings, Annual Conference Series ACM SIGGRAPH, 55-64 (July, 2000). With Fabio Pellacini, and James A. Ferwerda.
"A Lab Ahead of its Time: Cornell Graphics Lab Sets High Standards." Architectural Record, 198-204 (June, 2000). With B. J. Novitski, and Moreno A. Piccolotto.
"Approximate Visibility for Illumination Computations using Point Clouds." Program of Computer Graphics Technical Report, Cornell University (June 1, 2000). With Philip M. Dutre, and Parag Tole.
"A System for 3D Conceptual Modeling for Architectural Design." Program of Computer Graphics Technical Report, Cornell University (January 3, 2000). With Moreno Piccolotto, Sebastian Fernandez, Kavita Bala, and Michael Malone.
"Interactive Direct Lighting in Dynamic Scenes." Program of Computer Graphics Technical Report, Cornell University (January 2, 2000). With Sebastian Fernandez, Kavita Bala, and Moreno Piccolotto.

Zygmunt Haas
Associate Professor
Member of the School of Electrical and Computer Engineering
and the Graduate Field of Computer Science
haas@cs.cornell.edu
http://www.ee.cornell.edu/$\sim$haas/wnl.htm
Ph.D. Stanford University

My research is in the area of mobile and wireless systems and networks. Selected examples of the projects that are conducted in my Wireless Network Laboratory (WNL) are: ad-hoc networks (routing, medium access control, security, etc.), quality of service, cross-layer protocol design, mobile web access, multicasting, and mobility management.

Memberships
Professional Societies: IEEE Communications Society; IEEE Vehicular Technology; Association for Computing Machinery (ACM); Special Interest Group on Mobile Communications (SIGMOBILE).

Awards/Honors
Michael Tien '72 Award, Cornell College of Engineering, Excellence in Teaching Award (September, 2000).

University Activities
Member, Ad-hoc Tenure Promotion Review Committee, Computer Science Department, Cornell University;
      Computing Policy Committee (CPC), College of Engineering, Cornell University.
EE Policy Committee, School of Electrical and Computer Engineering, Cornell University (2000).
Member of the Committee on "Bits On Our Minds 2001 (BOOM '01).

Professional Activities
Editorial Board: IEEE Transactions on Networking.
Editorial Board: IEEE Communications Magazine.
Organizer and chair of the session on "Outrageous Opinions" at MobiHoc '01, The ACM Symposium on Mobile
      Ad-hoc Networking & Computing, Long Beach, CA (October 4-5, 2001).
Member: IASTED (The International Association of Science and Technology for Development) Technical
      Committee on Telecommunication.
Member:ACM Mobicom Steering Committee.
Chair of the IEEE Technical Committee on Personal Communications (TCPC).
Vice-chair of the IEEE Technical Committee on Personal Communications (TCPC).
Editorial Board of the journal Wireless Communications and Mobile Computing, John Wiley & Sons.
Guest Editor: Wireless Personal Communication Journal, special issue on Multimedia Network Protocols and
      Enabling Radio Technologies.
Editorial Board: ACM/Baltzer Wireless Networks.
Editorial Board: Journal of High Speed Networks.
Program Committee: IEEE Symposium on Ad-hoc Wireless Networks (SAWN), in conjunction with IEEE
      GLOBECOM 2001, San Antonio, TX (November 25-29, 2001).
Program Committee member and session chair, Milcom'01, McLean, VA (October 28-31, 2001).
Committee member: Wireless Communications and Networking Conference 2002 (WCNC'02), Orlando, FL
      (March 18-21, 2002).
Program Committee: European Wireless 2002, Next Generation Wireless Networks: Technologies, Protocols,
      Services and Applications, Florence, Italy (February 26-28, 2002).
Program Committee: 11th IEEE Workshop on Local and Metropolitan Area Networks, Boulder, CO (May 18,
      2001).
Program Committee: IEEE Wireless Networks and Mobile Computing Workshop, Phoenix, AZ (April 16-19, 2001).
Program Committee: ACM/IEEE MobiCom'2000, Boston, MA (August 6-11, 2000).
Program Committee: First IEEE Workshop on Mobile Ad HOC Networking and Computing Workshop
      (MobiHOC), Boston, MA (August 11-12, 2000).
NSF reviewer and panelist.

Lectures
Research in the Wireless Networks Laboratory at Cornell. Department of Electrical Engineering, Columbia
      University (April 23, 2001).
Research in the Wireless Networks Laboratory at Cornell. Communication Systems Department, Swiss
      Federal Institute of Technology Lausanne (EPFL) (December 11, 2000).

Publications
"The Zone Routing Protocol: A Hybrid Framework for Routing in Ad-hoc Networks." Ad-hoc Networks, Charlie
      Perkins, editor. Addison Wesley (2001). With M. R. Pearlman.
"A Communication Infrastructure for Smart Environments - A Position Article." IEEE Personal Communications
      Magazine, special issue on "Networking the Physical World," 6-10 (October , 2000).
"Predictive Distance-based Mobility Management for PCS Networks." 2000 Cornell Summer Workshop on
      Information Theory (Bergerfest), Ithaca, NY (August 18-19, 2000). And B. Liang.
"On the Impact of Alternate Path Routing for Load Balancing in Mobile Ad-hoc Networks." First Annual
      IEEE/ACM Workshop on Mobile Ad-hoc Networking & Computing, MobiHOC'2000, Boston, MA (August
      11, 200). With M. R. Pearlman, P. Scholander, and S. S. Tabrizi.
"A Decision-theoretic Approach to Resource Allocation in Wireless Multimedia Networks." Fourth International
      Workshop on Discrete Algorithms and Methods for Mobile Computing, DIALM 2000, Boston, MA (August
      11, 2000). And J. Y. Halpern, L. Li, and S. B. Wicker.
"Securing Ad-hoc Networks." IEEE Network Magazine 13(6) (November/December 1999) With L. Zhou.
"Determining the Optimal Configuration for the Zone Routing Protocol." IEEE Journal on Selected Areas in
      Communications (JSAC), issue on Ad-hoc Networks 17(8):1395-1414 (August, 1999). And M. R.
      Pearlman.
"The Dynamic Packet Reservation Multiple Access Scheme for Multimedia Traffic." ACM/Baltzer Mobile
      Networks Applications 4:87-99 (1999). With D. A. Dyson.
"Ad-hoc Location Management using Quorum Systems." IEEE Transactions on Networking, ACM/IEEE
      Transactions on Networking (April, 1999). And B. Liang.
"The Multiply-detected Macrodiversity Scheme for Wireless Cellular Systems." IEEE Transactions on
      Vehicular Technology 47(2) (May, 1998). And C-P. Li.

New Patent Applications
"ITAMAR - Independent Tree Ad-hoc Multicast Routing." Cornell Research Foundation, Application number: D-
      2823 - Haas. And M. S. Sajama.
"COCA: A Secure Distributed On-line Certification Authority." Cornell Research Foundation, Application
      number: D-2732A - Haas. With F. Schneider, L. Zhou, and R. van Renesse.
"Adaptive Power Control in Wireless Ad-hoc Networks." Cornell Research Foundation, Application number: D-
      2507 - Haas. And Miguel Sanchez.
"Routing and Mobility Management Protocols for Ad-hoc Networks." Cornell Research Foundation, Application
      number: D-2191 - Haas.

Joseph Halpern
Professor, and Co-director:
Cognitive Studies Program
halpern@cs.cornell.edu
http://www.cs.cornell.edu/home/halpern/
Ph.D. Harvard University, 1981

My research is concerned with representing and reasoning about knowledge and uncertainty in multi-agent systems. The work uses tools from logic (particularly modal logic and the idea of possible-worlds semantics), probability theory, distributed systems, game theory, and AI, and I like to think that it contributes to our understanding of each of these areas as well.
     Some themes of my current research include: (1) applying ideas of decision theory to constructing algorithms in asynchronous distributed systems, database systems, and wireless systems, (2) providing foundations for useful qualitative notions of decision theory, (3) reasoning about security.

Honors
Guggenheim Fellowship (for 2001-02).
Fulbright Fellowship (for 2001-02).

University Activities
Co-director: Cognitive Studies Program.
Chair, Admissions Committee, Department of Computer Science.

Professional Activities
Fellow, American Association of Artificial Intelligence.
Editor-in-chief: Journal of the ACM (as of May, 1997).
Consulting Editor: Chicago Journal of Computer Science.
Editorial board: Artificial Intelligence Journal; Information and Computation; Journal of Logic and Computation.
Member: ACM Publications Board.
Chairman: ACM Preprint Repository.
Coordinator: CoRR (Computing Research Repository).
Member: LICS (IEEE Conference on Logic in Computer Science) Advisory Board.
President of Board of Directors: Corporation for Theoretical Aspects of Reasoning About Knowledge.
Program Chair, 16th Annual IEEE Symposium on Logic in Computer Science (2001).
Program Committee Member and Conference Chair, 8th Conference on Theoretical Aspects of Rationality and
      Knowledge (2001).

Lectures
A Computer Scientist Looks at Game Theory.
-. Invited talk, Games 2000, Bilbao, Spain (July, 2000).
-.Invited talk, 4th Conference on Logic and Foundations of Game and Decision Theory, Torino, Italy (July,
      2000).
Plausibility Measures and Default Reasoning. Amsterdam University (May, 2001).
Plausibility Measures: a General Approach for Representing Uncertainty Northwestern University (May, 2001).
Complexity, Logic, and Computation: A Symposium in Honor of Albert Meyer, Boston, MA (June, 2001).

Publications
"Axiomatizing Causal Reasoning." Journal of AI Research 12:317-337 (2000).
"A Note on Knowledge-based Programs and Specifications." Distributed Computing 13(3):145-153 (2000).
"First-order Conditional Logic Revisited." ACM Transactions on Computational Logic 1(2):175-207 (2000). With
      N. Friedman and D. Koller.
"Multi-agent Only Knowing." Journal of Logic and Computation 11(1):41-70 (2001). With G. Lakemeyer.
"A Logic for SDSI's Linked Local Named Spaces." Journal of Computer Security 9(1,2):47-74 (2001). With R.
      van der Meyden.
"A Decision-theoretic Approach to Reliable Message Delivery. Distributed Computing 14:1-16 (2001). With F.
      Chu.
"A Decision-theoretic Approach to Resource Allocation in Wireless Multimedia Networks." Proceedings of Dial
      M for Mobility, 86-95 (2000). With Z. Haas, L. Li, and S. B. Wicker.
"Conditional Plausibility Measures and Bayesian Networks." Proceedings of the Sixteenth Conference on
      Uncertainty in AI, 247-255 (2000).
"Minimum-energy Mobile Wireless Networks Revisited." Proceedings of the IEEE Conference on
      Communications (2001). With L. Li.
"A Logical Reconstruction of SPKI." Proceedings of the 14th IEEE Computer Security Foundations Workshop
      (2001). With R. van der Meyden.
"Review of 'Probability and Conditionals: Belief Revision and Rational Decisions.'" Journal of Philosophical
      Logic 100(2):277-281 (2000).
"The Unusual Effectiveness of Logic in Computer Science." Bulletin of Symbolic Logic 7(2):213-236 (2001).
      With R. Harper, N. Immerman, P. G. Kolaitis, M. Y. Vardi, and V. Vianu.
"Editorial: An Author's Bill of Rights and Responsibilities." Journal of the ACM 47(5):828-825 (2000).
"CoRR: A Computing Research Repository (with commentary)." ACM Journal of Computer Documentation
      24(2):41-48 (2000).

Landmark Publications
Reasoning About Knowledge. MIT Press (1995). With R. Fagin, Y. Moses, and M. Y. Vardi.
"Knowledge and Common Knowledge in a Distributed Environment." Journal of the ACM 37(3):549-587 (1990).
      With Y. Moses. Awarded Godel Prize in 1997.
"Belief, Awareness, and Limited Reasoning." Artificial Intelligence 34:39-76 (1988). With R. Fagin. Conference
      version winner of MIT Press Publisher's Prize as best paper of the 9th International Joint Conference on
      Artificial Intelligence (1985).
"An Analysis of First-order Logics of Probability." Artificial Intelligence 46:311-350 (1990). Conference version
      winner of Publisher's Prize as best paper of the 11th International Joint Conference on Artificial Intelligence
      (1989).
"Plausibility Measures and Default Reasoning." Proceedings of the Thirteenth National Conference on Artificial
      Intelligence, 1297-1304 (1996). To appear in the Journal of the ACM . With N. Friedman. Commended for
      its excellence by the Committee on the "IGPL/FoLLI Prize for the Best Idea of the Year 1996."

Juris Hartmanis
Walter R. Read Professor of Engineering
Turing Award Winner
jh@cs.cornell.edu
Ph.D. California Institute of Technology, 1955

The strategic goal of my research is to contribute to the development of a comprehensive theory of computational complexity. Computational complexity, the study of the quantitative laws that govern computation, is an essential part of the science base needed to guide, harness, and exploit the explosively growing computer technology. My current research interests focus on understanding the computational complexity of chaotic systems and the classification of undecidable problems in complexity theory.

Awards
Recipient of the Grand Medal of the Latvian Academy of Sciences (Lielo Medalu) (2001).

Professional Activities
Member: National Academy of Engineering.
Foreign Member: Latvian Academy of Sciences.
Fellow: American Academy of Arts and Sciences; New York State Academy of Sciences; AAAS.
Editor: Springer-Verlag Lecture Notes in Computer Science; Journal of Computer and Systems Sciences; Fundamenta Informaticae.
Advisory Board: EATCS Monographs in Theoretical Computer Science; Springer-Verlag; International Journal
      of Foundations of Computing.
Member: DIMACS External Advisory Committee.
Member: Santa Fe Institute Science Board.
Member: Santa Fe Institute Science Steering Committee.
Member: University of Cincinnati Computing Program Review Committee (November 1-3, 2000).

University Activities
Chair: Engineering College Nominating Committee.
Member: FCI Founders Committee.

Lectures
Goedel, Undecidability and Automata Theory, Half Century of Automata Theory. University of Western Ontario
       (July 26, 2000).
Four lectures at Jyvaskyle University, Finland, (August 10-11, 2000).
-. Undecidability and Incompleteness Results in Theory of Computing
-. Succinctness and Minimality of Automata Description
-. Search for Limits of Feasible Computations
-. On the Complexity and Shape of Mathematical Proofs
Computational Complexity and Mathematical Proofs, University of Saarbruecken, Germany (August 30, 2000).
What can Computational Complexity Theory Say about Mathematics?
-. Iowa State University (October 23, 2000).
-. Cray Lecture Series, University of Minnesota (October 30, 2000).

Publications
"Computational Complexity and Mathematical Proofs." Informatics- 10 Years Back, 10 Years Ahead.
      Reinhard Wilhelm, editor. Springer-Verlag LNCS 2000, 251-256 (2001).

Mark Heinrich
Assistant Professor
Member of the School of Electrical and Computer Engineering
and the Graduate Field of Computer Science
heinrich@csl.cornell.edu
http://www.csl.cornell.edu/~heinrich/
Ph.D. Stanford University, 1998

My research is concerned with the design of active memory and I/O systems for next-generation servers and data-intensive computing. This work has focused on extending the cache coherence mechanism in modern servers to implement active memory operations-computation performed in the memory system on behalf of the microprocessor to speed up overall execution time. Coupled with this work is the exploration of the effect of new networking technologies (i.e. InfiniBand) on next-generation servers and the integration of active memory and I/O techniques with this networking technology. We have also shown that active memory machines and hardware cache-coherent distributed shared-memory (DSM) machines need much the same support, and, in fact, that by building our single-node active memory system we can also support a multiprocessor version of the machine that we call active memory clusters. Active memory clusters can achieve hardware DSM performance at the low cost of clusters.
     In my work on Active I/O systems I am developing a smart InfiniBand switch (which can also be used in active memory clusters) that can support either normal or intelligent I/O devices, and offload computation from the microprocessor to minimize latency and reduce bandwidth requirements in the I/O system. This work also involves innovative operating system restructuring, including the filesystem and the network stack. The operating system in active I/O systems must be partitioned between the microprocessor and the active I/O devices. Our own operating system, SplitOS, is joint work between our research group and groups at Rutgers and Princeton.
     In work on scalable cache coherence protocols, I am working on issues of fairness and robustness in scalable distributed shared-memory (DSM) machines. In addition, I am looking at the quantitative impact of many coherence protocol techniques by evaluating each technique on a flexible DSM prototype. Together with Martin Burtscher, I am also exploring predictive techniques in cache coherence protocols to minimize latency.

Awards/Honors
Michael Tien '72 Excellence in Teaching Award (2000-2001).
IEEE Teacher of the Year Award (1999-2000).
NSF Faculty Early Career Development Award (2000-2004).

University Activities
Member: Intelligent Information Systems Institute; ECE Curriculum and Standards Committee; ECE Long-
      Range Recruiting Committee; ECE Experimental Systems Recruiting Committee; ECE Circuits & MEMS
      Recruiting Committee; CURIE Summer Program for Women in Engineering; Fields of Electrical
      Engineering, Computer Science.

Professional Activities
Publicity Chair: International Symposium on High-performance Computer Architecture (PCA) (January, 2001).
Panelist: Mathematics, Information, and Computational Sciences Division within the Office of Advanced
      Scientific Computing Research in the Office of Science at the U.S. Department of Energy (DOE) (April,
      2001).
Program Committee: Workshop on Caching, Coherence, and Consistency (WC3), held in conjunction with the
      ACM Conference on Supercomputing (June, 2001).
Program Committee: International Conference on Parallel Architectures and Compilation Techniques (PACT)
      (September, 2001).

Lectures
Flash Forward: Better, Faster, Cooler. Cornell University Silicon Valley Event, hosted by Hunter Rawlings, San
      Mateo, CA (April, 2001).
Providing Hardware DSM Performance at Software DSM Cost. Seminar, University of Rochester (April, 2001).
Hardware DSM Performance at Software DSM Cost. Air Force Research Laboratory, Rome, NY (March, 2001).
A Case for Asynchronous Active Memories. ISCA Workshop on Solving the Memory Wall (June, 2000).

Publications
"FLASH vs. (Simulated) FLASH: Closing the Simulation Loop." In Proceedings of the 9th International
      Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), 49-
      58 (November, 2000). With J. Gibson, R. Kunz, D. Ofelt, M. Horowitz, and J. Hennessy.
"Using Meta-level Compilation to Check FLASH Protocol Code." In Proceedings of the 9th International
      Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), 59-
      70 (November, 2000). With A. Chou, B. Chelf, and D. Engler.
"A Case for Asynchronous Active Memories." In Proceedings of the ISCA Workshop on Solving the Memory
      Wall (June, 2000). With R. Manohar.

Sheila Hemami
Associate Professor
Member of the School of Electrical and Computer Engineering
and the Graduate Field of Computer Science
hemami@cs.cornell.edu
http://www.ece.cornell.edu/people/faculty/faculty_list.shtml#shei
Ph.D. Stanford University, 1994

The emerging information superhighway provides an example of the flexibility required of image and video compression and transmission techniques. Varying network capacities, differences in viewing devices, and a broad spectrum of user needs suggest the desirability of coding techniques that can efficiently span large quality and bandwidth ranges. Additionally, coded data must be robust to errors and loss of varying degrees across multiple network segments. For practicality, algorithms must be inexpensive to implement, in either hardware or software. My research interests broadly concern such communication of visual information. Particular topics of interest include multirate video coding and transmission, compression specific to packet networks and other lossy networks, and psychovisual considerations.

University Activities
College of Engineering Committee of Faculty Women.
Society of Women Engineers Faculty Advisor.
EE Curriculum and Standards Committee.
Peoplesoft Lab Oversight Committee.
EE General Recruiting Committee.
Summer presentation to students in "Inventing an Information Society" (July, 2000).

Professional Activities
Reviewer, IEEE Trans. Signal Processing; IEEE Trans. Image Processing, IEEE Trans. Circuits and Systems for Video Technology; IEEE Communications Letters; IEEE ICIP 2001.
Associate Editor, IEEE Trans. Signal Processing.

Awards/Honors
HKN C. Holmes MacDonald Outstanding Teaching Award.
Michael Tien '72 Cornell College of Engineering Teaching Award.
Fulbright Distinguished Lecturer, State of Morocco (2001).

Publications
"Perceptual Quantization for Wavelet-based Image Coding." Proc. IEEE Int. Conf. on Image Processing,
      Vancouver, B.C. (September, 2000). With M. G. Ramos.
"Subjective Quality Evaluation of Low Bit Rate Video." Proc. Human Vision and Electronic Imaging 2001, San
      Jose, CA (January, 2001). With M. A. Masry, W. Osberger, and A. M. Rohaly.
"Generalized Rate-distortion Optimizations for Motion-compensated Video Coding." IEEE Trans. Circuits &
      Systems for Video Tech. (September, 2000). With Yan Yang.
"Perceptually-based Robust Image Transmission over Wireless Channels." Proc. IEEE Int. Conf. on Image
      Processing, Vancouver, B.C. (September, 2000). With M. E. Buckley, M. G. Ramos, and S. B. Wicker.
"Robust Data Hiding using Psychovisual Thresholding." Proc. IEEE Int. Conf. on Image Processing,
      Vancouver, BC, Canada (September, 2000). With M. A. Masry, and M. G. Ramos.

John Hopcroft
Joseph Silbert Dean of Engineering
Turing Award Winner
Professor
jeh17@cornell.edu
http://www.cs.cornell.edu/jeh
Ph.D. Stanford University, 1964

Since January 1994, I have been the Joseph Silbert Dean of the College of Engineering. Upon completion of my term as dean, which ends on June 30, 2001, I plan to return to research in the Department of Computer Science at Cornell. My research will center on the study of information capture and access. I have also been involved in the theoretical aspects of computing, especially analysis of algorithms, formal languages, automata theory, and graph algorithms. I have coauthored four books on formal languages and algorithms with Jeffrey D. Ullman and Alfred V. Aho.
     I was formerly the associate dean for college affairs and the Joseph C. Ford Professor of Computer Science. After receiving an M.S. (1962) and Ph.D. (1964) in electrical engineering from Stanford University, I spent three years on the faculty of Princeton University. In 1967, I joined the Cornell faculty, was named professor in 1972 and served as chairman of the Department of Computer Science from 1987 to 1992. I am an undergraduate alumnus of Seattle University, and was honored with a Doctor of Humanities Degree, Honoris Causa, in 1990.

Professional Activities
National Academy of Engineering.
Scientific Advisory Committee for The David and Lucile Packard Foundation;
John Von Neumann Medal Selection Committee for the Institute of Electrical and Electronics Engineers, Inc.
      (IEEE);
School of Engineering Advisory Committee, Hong Kong University of Science and Technology;
National Advisory Committee on Informatics Engineering, National College of Industrial Relations (Ireland).
Fellow: American Academy of Arts and Sciences; American Academy for the Advancement of Science; Institute of Electrical and Electronics Engineers (IEEE); (Charter) ACM.
Associate Editor (and Editor): Journal of Computer and System Sciences.
Editor (and Member, Executive Committee): Algorithmica.
Associate Editor: Information Sciences.
Editor: International Journal of Computational Geometry and Applications.

Klara Kedem
Professor
kedem@cs.cornell.edu
http://www.cs.cornell.edu/kedem
Ph.D. Tel Aviv University, 1989

My research area is Computational Geometry with applications to problems in computer vision and bio-information. The attempt to deal with practical problems (like shape comparison) by investigating their geometric nature, yields a better theoretical understanding of the problems and provides sound and efficient algorithms. Among the theoretical problems I work on are problems in geometric optimization, such as covering a set of points by a given number of shapes and facility location. I have investigated the minimum Hausdorff distance as a tool for measuring shape resemblance between images.
    Many practical problems in the area of shape comparison seek a fully automated solution. The robustness of the minimum Hausdorff distance lends itself to such problems. In the last two years I have looked into shape comparison problems in three dimensions. Such problems arise in many life-science disciplines. In computational molecular biology I have come up with a new metric, the URMS, to measure substructure resemblance between proteins. This measure has been implemented and further applied to the analysis of molecular dynamics. It proves superior to previously used measures. With the department of life sciences at Ben Gurion University, I have been looking at dendrite shape comparison and classification. Here we applied a three dimensional Hausdorff distance for the problem and are in the midst of devising new measures.

Professional Activities
Chair of the Computer Science Department, Ben-Gurion University of the Negev, Israel.
Editorial board: Pattern Recognition Society Journal.
Guest editor of Computational Geometry: Theory and Applications.
Workshop program committee: Workshop on Algorithmic Engineering (WAE 01) BRICS, University of Aarhus,
      Denmark (August 28-31, 2001).
Reviewer: Israeli-U.S. Bi-national Science Foundation; The Academia - Ministry of Science, Israel; Pattern Recognition Journal; International Journal of Computational Geometry and Applications.

Grants:
LSRT Consortium, "Optimal Layout Design of Large Scale Satellite Telephony Systems in Rural Regions"
      (2000-2001).
Israeli Defense Ministry, "BGU Bioinformatics Center for the Interpretation of the Human Genome" (2001-2002).

Publications
"Optimal Facility Location under Various Distance Functions." Accepted for publication in International Journal
      of Computational Geometry and Applications. With S. Bespametnik, M. Segal and A. Tamir.
"Improved Algorithms for Placing Undesirable Facilities." Accepted for publication in Computers and Operations Research. With M. J. Katz, and M. Segal.

Jon Kleinberg
Associate Professor
kleinber@cs.cornell.edu
http://www.cs.cornell.edu/home/kleinber/
Ph.D. MIT, 1996

My research is concerned with algorithms that exploit the combinatorial structure of networks and information.
     The information we deal with is taking on an increasingly networked structure; the World Wide Web serves as perhaps the most compelling example of this phenomenon. One direction I have pursued, motivated by this issue, is the development of algorithms that analyze the link structure of the Web to identify high-quality information resources - hubs and authorities relevant to broad topics. A second direction is the development of graph models that can provide insight into the structure of large networks. I have been studying an algorithmic framework in which to frame questions about certain 'small-world' properties of networks; and with Duncan Callaway, John Hopcroft, Mark Newman, and Steve Strogatz, I have recently looked at models of random graphs that evolve over time. A third line of research has been concerned with designing algorithms that facilitate the rapid spread of information in a network; with David Kempe and Al Demers, I have worked on the design of randomized 'gossip' protocols that provide efficient, decentralized mechanisms for such tasks.
     Discrete optimization provides a collection of powerful techniques for approaching problems in this area; in particular, balancing constraints imposed by many individuals leads to a range of interesting new problems. With Christos Papadimitriou and Prabhakar Raghavan, I have been studying an optimization-based framework for clustering and data mining in which one has access to data from a large population; we consider the extent to which designing effective algorithms can be balanced with concerns about the privacy of each individual's data. With Amit Kumar, I have been looking at resource allocation problems with multiple users; since each user wants as much of the resource as possible, there are many competing objectives, and it is natural to seek the 'fairest' solution.
     Finally, many of the core problems in computational biology require algorithms for analysis of large volumes of data; such data is being generated at an accelerating pace by experimental studies of genomes and proteins across a wide range of species. With Debra Goldberg, Susan McCouch, and David Liben-Nowell, I have been investigating mathematical and computational approaches to the analysis of evolutionary relationships among species at the genomic level and the connection between this type of analysis and the problem of comparative mapping. I have also developed algorithms for relating sequence information to protein structure, and with Ron Elber I am continuing to look at approaches to this issue based on threading methods.

Awards
National Academy of Sciences Award for Initiatives in Research (2001).
David and Lucile Packard Foundation Fellowship (1999-2004).
ONR Young Investigator Award (1999-2002).
NSF Faculty Early Career Development Award(1997-2001).
Alfred P. Sloan Research Fellowship (1997-1999).

University Activities
Member: Cornell Faculty of Computing and Information (FCI) Founders.

Professional Activities
Member: National Academies Computer Science and Telecommunications Board study on Fundamentals of
      Computer Science.
Program Committees: IEEE Symposium on Foundations of Computer Science, 2001; International World Wide
      Web Conference (2001); ACM International Conference on Knowledge Discovery and Data Mining (2001); Workshop on Algorithms and Data Structures (2001); Workshop on Randomization and Approximation Techniques in Computer Science (2001).
Lecturer: Lipari Summer School on On-line Algorithms (2000).
NSF Review Panel member.

Lectures
Information Networks: Models and Algorithms.
-. America Online (December, 2000).
-. Packard Fellowship Annual Meeting (September, 2000).
-. Santa Fe Institute Workshop on Complex Interactive Networks (August, 2000).
Small-World Phenomena and the Dynamics of Information.
-. Workshop in Honor of Allan Borodin's 60th Birthday (June, 2001).
-. Snowbird Learning Workshop (April, 2001).
Structure and Content in World-Wide Web Search. Invited plenary lecture at the Meeting of the North American Chapter of the Association for Computational Linguistics (June, 2001).
Detecting a Network Failure. IEEE Symposium on Foundations of Computer Science (November, 2000).
On-Line Algorithms for Routing and Search. Lipari Summer School (July, 2000).

Publications
"Spatial Gossip and Resource Location Protocols." Proc. 33rd ACM Symposium on Theory of Computing
      (2001). With D. Kempe and A. Demers.
"Provisioning a Virtual Private Network: A Network Design Problem for Multicommodity Flow." Proc. 33rd ACM
      Symposium on Theory of Computing (2001).With A. Gupta, A. Kumar, R. Rastogi, and B. Yener.
"On the Value of Private Information." Proc. 8th Conf. on Theoretical Aspects of Rationality and Knowledge
      (2001). With C. Papadimitriou, and P. Raghavan.
"Adversarial Queuing Theory." Journal of the ACM 48(1):13-38 (2001). With A. Borodin, P. Raghavan, M.
      Sudan, and D. Williamson.
"Universal-stability Results and Performance Bounds for Greedy Contention-resolution Protocols." Journal of
      the ACM 48(1):39-69 (2001). With D. M. Andrews, B. Awerbuch, A. Fernandez, F. T. Leighton, and Z. Liu.
"Navigation in a Small World." Nature 406:845 (2000).
"Detecting a Network Failure." Proc. 41st IEEE Symposium on Foundations of Computer Science, 231-239
      (2000).
"Fairness Measures for Resource Allocation." Proc. 41st IEEE Symposium on Foundations of Computer
      Science, 75-85 (2000). With A. Kumar.
"Algorithms for Constructing Comparative Maps." Comparative Genomics: Empirical and Analytical
      Approaches to Gene Order Dynamics, Map Alignment and the Evolution of Gene Families (David Sankoff
      and Joseph H. Nadeau, editors), 243-261 (2000). With D. Goldberg and S. McCouch.
"The Syntenic Diameter of the Space of N-chromosome Genomes." Comparative Genomics: Empirical and
      Analytical Approaches to Gene Order Dynamics, Map Alignment and the Evolution of Gene Families
      (David Sankoff and Joseph H. Nadeau, editors), 185-198 (2000). With D. Liben-Nowell.
"Allocating Bandwidth for Bursty Connections." SIAM J. Computing 30(1):191-217 (2000). With Y. Rabani, and
      E. Tardos.
"Node-disjoint Paths on the Mesh, and a New Trade-off in VLSI Layout." SIAM J. Computing 29(4):1321-1333
      (2000). With A. Aggarwal, and D. Williamson.
"Approximation Algorithms for Classification Problems with Pairwise Relationships: Metric Labeling and
      Markov Random Fields." Proc. 40th IEEE Symposium on Foundations of Computer Science, 14-23 (1999).
      With E. Tardos.
"Efficient Algorithms for Protein Sequence Design and the Analysis of Certain Evolutionary Fitness
     Landscapes." Journal of Computational Biology 6(3-4):387-404 (1999).
"Authoritative Sources in a Hyperlinked Environment." Journal of the ACM 46(5): 604-632 (September,1999).

Dexter Kozen
Joseph Newton Pew, Jr.
Professor of Engineering
kozen@cs.cornell.edu
http://www.cs.cornell.edu/kozen/
Ph.D. Cornell University, 1977

My research interests include the theory of computational complexity, especially complexity of decision problems in logic and algebra, program logic and semantics, and computational algebra. Recent work includes new polynomial-time algorithms for type inference in type systems with subtypes and recursive types; algorithms solving systems of set constraints as used in program analysis; a unification algorithm for set constraints and a new constraint logic programming language based on set constraints; development of the theory of rational spaces and their relationship to set constraints; an algorithm for decomposition of algebraic functions; a new polynomial-time algorithm for resolution of singularities of plane curves; efficient algorithms for optimal transmission of encoded video data; optimality results for digital interleavers; and expressiveness, complexity and completeness results for Kleene algebras with tests. Recently I have begun to investigate algorithms for efficient code certification and the application of Kleene algebra with tests to the verification of compiler optimizations.

Honors
Class of 1960 Scholar, Williams College.
Stephen and Margery Russell Distinguished Teaching Award, College of Arts and Sciences, Cornell.

University Activities
Member: Undergraduate Admissions Committee, College of Engineering; University Arbitration Panel.
Faculty Advisor: Cornell Men's Rugby Football Club; Johnson Graduate School of Management Rugby Football
      Club; Cornell Women's Rugby Football Club.

Professional Activities
Program Committee Member: Foundations of Software Science and Computation Structure; Mathematical
      Foundations of Computer Science.
Editorial Board Member: Journal of Relational Methods in Computer Science; Theory of Computing Systems.
Supervisory Board: Centre for Basic Research in Computer Science (BRICS), Aarhus University; Goedel Prize
       Committee.

Lectures
Language-based Security. Department of Computer Science, Dartmouth College, Hanover, NH (March, 2000).
On the Completeness of Propositional Hoare Logic. RelMiCS 5 Conference, Quebec City, Canada (January,
      2000).
A Computer Scientist's View of Admissible Sets. Tarski Centenary Conference, Warsaw, Poland (May, 2001).

Publications
"Certification of Compiler Optimizations using Kleene Algebra with Tests." In Proc. 1st Int. Conf. Computational
      Logic (CL2000), v. 1861 of Lecture Notes in Artificial Intelligence, J. Lloyd, V. Dahl, U. Furbach, M. Kerber,
      K.-K. Lau, C. Palamidessi, L. M. Pereira, Y. Sagiv, and P. J. Stuckey, editors, London, Springer-Verlag,
      568-582 (July 2000).
"On the Completeness of Propositional Hoare Logic." Proc. of the 5th International Seminar Relational
      Methods in Computer Science (RelMiCS 2000), 195-202 (January, 2000). With J. Tiuryn.
"On Hoare Logic and Kleene Algebra with Tests." Trans. Computational Logic 1(1):60-76 (July, 2000).
"A Note on the Complexity of Propositional Hoare Logic." Trans. Computational Logic 1(1):171-174 (July,
      2000). With E. Cohen.
Dynamic Logic. MIT Press, Cambridge, MA (2000). With D. Harel and J. Tiuryn.
"Myhill-Nerode Relations on Automatic Systems and the Completeness of Kleene Algebra." In Proc. 18th
      Symp. Theoretical Aspects of Computer Science, A. Ferreira and H. Reichel, editors. Dresden, Germany
      (February, 2001). Lecture Notes in Computer Science, v. 2010, Springer-Verlag, 27-38.
"Intuitionistic Linear Logic and Partial Correctness." Proc. 16th Symp. Logic in Computer Science. IEEE
      (June, 2001) With J. Tiuryn.

Dean Krafft
Senior Research Associate
Director of Computing Facilities
dean@cs.cornell.edu
http://www.cs.cornell.edu/home/dean
Ph.D. Cornell University, 1981

Iserve both as a researcher and an administrator in the Department of Computer Science at Cornell. In my guise as an administrator, I manage the Computing Facilities Support group and worry about a number of issues including computer security, networking, and building web services. Most recently, as part of the Nomad research project (http://www.nomad.cornell.edu), I have been leading a campus-wide pilot of megabit-speed wireless LAN technology. This will be rolled out by Cornell Information Technologies as a Cornell campus service in the fall 2001.
    On the research side, I am part of the Cornell Digital Libraries Research Group (CDLRG - http://www.cs.cornell.edu/cdlrg). A major focus of our effort is on interoperability issues for digital libraries. As part of that broader thrust, I am a Co-principal Investigator on the NSF-funded National Science Digital Library Project at Cornell (http://www.SiteForScience.org). My own particular interests focus on ensuring the availability in the digital world of pre-digital published and manuscript materials, as well as related issues on copyright, the public domain, and public access to older and out-of-print materials.

Cristoph Kreitz
Senior Research Associate
kreitz@cs.cornell.edu
http://www.cs.cornell.edu/home/kreitz
Ph.D. FernUniversitaet Hagen, 1984

My primary research interest is the application of automated deduction to the design, verification, and optimization of software systems. My current research aims at developing a Logical Programming Environment for the construction of reliable and efficient distributed systems. In collaboration with the Nuprl and Ensemble research groups, I have built semantics-based tools for the automatic optimization of protocol stacks in the Ensemble group communication system. More recently, Robbert van Renesse, Mark Bickford and I have developed a generic switching protocol for the construction of adaptive systems and proved it correct with the Nuprl proof development system. For this purpose, we introduced the concept of meta-properties and used them to characterize communication properties that can be preserved by switching. We also identified switching invariants that an implementation of the switching protocol must satisfy in order to work correctly. The verification efforts revealed a variety of implicit assumptions that are usually made when designing communication systems and uncovered minor design errors that would otherwise have made their way into the implementation.
     I am also interested in the development of automatic proof procedures for classical and non-classical logics. Together with former students from the Technical University of Darmstadt, I work on proof search methods based on matrix-characterizations of logical validity, a very compact representation of the search space. We have developed a uniform proof search procedure for classical logic, intuitionistic logic, various modal logics, fragments of linear logic, and for inductive specification proofs. We have also developed a uniform algorithm for transforming the machine-found matrix proofs into sequent proofs. In the past year, we implemented JProver, a first-order intuitionistic theorem prover that creates sequent-style proof objects, and connected it as external proof engine to the interactive proof assistants Nuprl and MetaPRL. The combination of these systems gives a user the full expressive power of the proof assistant when dealing with complex proofs and verifications, while at the same time taking advantage of well-understood and efficient proof techniques for subproblems that only depend on first-order reasoning.

Professional Activities
Program Committee: IJCAR Workshop on Verification (2001).
Referee: Handbook of Automated Reasoning; Journal of Symbolic Computation; Journal of Functional Programming.

Lectures
The NuPRL Open Logical Environment. Scottish Theorem Proving Meeting, St. Andrews, Scotland (July,
      2000).
Matrix-based Inductive Theorem Proving, International Conference TABLEAUX-2000, St. Andrews, Scotland
      (July, 2000).
Building Reliable, High-performance Communication Systems from Components. Technical University of
      Darmstadt, Germany (July, 2000).
A Logical Programming Environment for Communication Systems. University of Potsdam, Germany
      (December, 2000).
Protocol Optimization in Ensemble/Spinglass. DARPA FTN PI Meeting, St. Petersburgh, (January, 2001).
Advances in Logical Programming Environments. DARPA PCES PI meeting, San Diego (February, 2001).
An Open Logical Programming Environment,.DARPA PCES PI meeting, St. Louis (May, 2001).
Formal Design of Reliable Software Systems. DARPA Information Survivability Conference and Exposition II,
      Anaheim (June, 2001).
Formal Design and Verification: Challenges and Prospects. Invited talk, workshop on verification, International
      Joint Conference on Automated Reasoning, Siena, Italy (June, 2001).

Publications
"J. Prover: Integrating Connection-based Theorem Proving into Interactive Proof Assistants." Proc. International
      Joint Conference on Automated Reasoning, LNAI, Springer (June 2001). With S. Schmitt, L. Lorigo, and A.
      Nogin.
"An Experiment in Formal Design using Meta-properties." Proc. DARPA Information Survivability Conference
      and Exposition II (DISCEX 2001), IEEE Computer Society Press (June 2001). With M. Bickford, R. van
      Renesse, and R. Constable.
"Protocol Switching: Exploiting Meta-properties." Proc. International Workshop on Applied Reliable Group
      Communication (WARGC 2001), IEEE Computer Society Press (April 2001). With X. Liu, R. van Renesse,
      M. Bickford, and R. Constable.
"A Uniform Procedure for Converting Matrix Proofs into Sequent-style Systems." Journal of Information and
      Computation 162 (1-2):226-254 (2000). With S. Schmitt.
"Matrix-based Inductive Theorem Proving." Proc. International Conference TABLEAUX-2000, LNAI 1847,
      Springer (July 2000). With B. Pientka.

Carl Lagoze
Digital Library Scientist
lagoze@cs.cornell.edu
http://www.cs.cornell.edu/lagoze/lagoze.html
MSE Wang Institute

Our group investigates the policies, organization, and architecture of distributed information spaces. The Web, and the massive amount of content that it makes available to us in our daily lives, provides the backdrop for our work. The goal of our research is to understand and prototype the services and organizational structures that we can build on top of this global information base in order to increase its functionality, integrity, and ease of use. We undertake this research with the recognition that any proposed solutions must balance the economy and speed of automated solutions against the often-irreplaceable expertise that comes from human intervention.
     Within this context we examine a number of research areas: Architectures for storage of and access to the multiple forms of digital content. Policies and enforcement mechanisms that facilitate the preservation and secure management of distributed content. The role of metadata, in its many forms, in the management of digital content. Protocols for federating information across distributed repositories and services. Architectures and services for interlinking amongst document references and citations.
      We work in close collaboration with other researchers in the Cornell, national and international library, computer science, and Internet communities. Our research model is highly applied: building standards and systems and supporting the deployment of them to our collaborators in the global information community. This research model has produced Dienst, an architecture and protocol for creating distributed document repositories and services, the Open Archives Initiative Metadata Harvesting Protocol that provides access to metadata in a variety of forms, and theFEDORA digital object model for managing access to highly variable digital content.

Professional Activities
Member: Dublin Core Metadata Initiative Advisory Council; Executive Committee, Open Archives initiative.
Program Committee: JCDL 2001; ECDL 2001; 10th International World Wide Web Conference.

Lectures
Business Unusual: How Event-awareness May Breathe Life into the Catalog. Bicentennial Conference on
      Bibliographic Control in the New Millennium. Library of Congress, Washington DC (November, 2000).
IFLA/DELOS/NSF Workshop: Standards and Metadata. EVA 2000 Moscow (November, 2000).
The Open Archives Initiative. Open Archives Initiative Open Days, Washington DC (January, 2001) and Berlin
      (February, 2001).
Metadata Musings. University of Virginia (April, 2001).

Publications
"The Open Archives Initiative: Building A Low-barrier Interoperability Framework." Joint Conference on Digital
      Libraries 2001, Roanoke, VA (September, 2001). With H. Van de Sompel.
"Combining RDF and XML Schemas to Enhance Interoperability Among Metadata Application Profiles." In 10th
      International World Wide Web Conference, Hong Kong (May 2001). With J. Hunter.
"Keeping Dublin Core Simple: Cross-domain Discovery or Resource Description." D-Lib Magazine (January,
      2001).
"Business Unusual; How 'Event Awareness' May Breathe Life into the Catalog." In Bicentennial Conference on
      Bibliographic Control in the New Millennium, Library of Congress, Washington DC (November, 2000).
"An Event-aware Model for Metadata Interoperability. In ECDL 2000, Lisbon (September, 2000). With J. Hunter,
      and D. Brickley.
"Policy-enforcing, Policy-carrying Digital Objects." In Fourth European Conference on Research and Advanced
      Technology for Digital Libraries, Lisbon (September, 2000). With S. Payette.

Lillian Lee
Assistant Professor
llee@cs.cornell.edu
http://www.cs.cornell.edu/home/llee
Ph.D. Harvard University, 1997

The ultimate goal of natural language processing is to enable computers to use human language as a communication medium both robustly and gracefully. However, because of the subtleties of human language, this goal cannot be achieved without access to large quantities of high-quality linguistic and domain knowledge. A major focus of my research is the development of "knowledge-lean" methods to overcome this knowledge acquisition bottleneck: I am investigating techniques that allow a system to automatically learn the requisite information directly from text.
     Recent work continues to explore the use of distributional similarity as a powerful tool for bootstrapping the knowledge acquisition process. The underlying idea is quite intuitive: information about an object can be gleaned from the objects that are similar to it, where similarity can be computed from unannotated data alone. But there are a multitude of ways to implement this idea; we are currently investigating different similarity-based paradigms, both theoretically and through experiments on large datasets.
     Also, work with Rie Kubota Ando has analyzed Latent Semantic Indexing, a commonly-used and highly influential information retrieval technique that seeks to uncover hidden document relationships without engaging in standard natural language processing. We have shown that the performance of this algorithm can be tied to the uniformity of the underlying topic-document distribution, and we have provided a new algorithm that automatically compensates for distributional non-uniformity.

Professional Activities
Program Chair: 2001 Conference on Empirical Methods in Natural Language Processing (EMNLP).
Area Chair (Corpus-based Natural Language Processing): Thirty-ninth Annual Meeting of the Association for
      Computational Linguistics (ACL).
Workshops Chair: Second Conference of the North American Chapter of the Association for Computational
      Linguistics (NAACL).
Editorial Board: Computational Linguistics (2000-2002).
Editorial Board: Machine Learning (2001-2004).
Program Committee: Second Conference of the North American Chapter of the Association for Computational
      Linguistics (NAACL) (2001).
Program Committee: Third Conference on Recent Advances in Natural Language Processing (RANLP) (2001).
Co-organizer: Text Learning: Beyond Supervision, a workshop at the Seventeenth International Joint
      Conference on Artificial Intelligence (IJCAI).
Invited Participant: Institute for Mathematics and its Applications (IMA) Workshop on Mathematical
      Foundations of Natural Language Modeling (2000).
Member: NSF review panel.
Technical coordinator: joint study agreement between Cornell University and IBM Watson on text mining and
      document management, 1999-present.
Affiliated faculty: Intelligent Information Systems Institute.
Referee: ACM Transactions on Information Systems.

University Activities
Department of Computer Science liaison to the Department of Linguistics.
Member: Women in Science and Engineering Advisory Group.
Member: Women in Science and Engineering Term Chair Award committee.
Reader: Arts and Sciences Undergraduate Admissions.
Member: Field of cognitive studies.
Affiliated faculty: Cornell Information Science program.

Lectures
Distributional Similarity: Models and Methods. Carnegie Mellon University (August, 2000).
Weakly-supervised Statistical Segmentation of Japanese. WhizBang! Labs (August, 2000).
Natural Language Technology (three-hour tutorial). Rome Air Force Lab, Knowledge Representation and
      Reasoning series (October, 2000).
Applications of EM Techniques. Institute for Mathematics and its Applications workshop on Mathematical
      Foundations of Natural Language Modeling (November, 2000).
On the Effectiveness of the Skew Divergence for Statistical Language Analysis. Eighth Meeting of Artificial
      Intelligence and Statistics (January, 2001).
Distributional Similarity and the Skew Divergence. Highland Technologies (April, 2001).
The Iterative Residual Rescaling Algorithm: An Analysis and Generalization of Latent Semantic Indexing.
      -. University of Maryland (April, 2001).
     -. Columbia University (May, 2001).

Publications
"Iterative Residual Rescaling: An Analysis and Generalization of LSI." Proceedings24th Annual International
      Conference on Research and Development in Information Retrieval (SIGIR) (2001). With Rie Kubota Ando.
"On the Effectiveness of the Skew Divergence for Statistical Language Analysis." Proceedings Artificial
      Intelligence and Statistics (2001).
"Mostly-unsupervised Statistical Segmentation of Japanese: Applications to Kanji." Proceedings First
      Conference of the North American Chapter of the Association for Computational Linguistics (NAACL)
      (2000). With Rie Kubota Ando.
"Measures of Distributional Similarity." Proceedings 37th Annual Meeting of the Association for Computational
      Linguistics (ACL) (1999).
"Similarity-based Models of Word Co-occurrence Probabilities." Machine Learning 34 (1999). With Ido Dagan,
      and Fernando Pereira.
"Fast Context-free Parsing Requires Fast Boolean Matrix Multiplication." Proceedings 35th Annual Meeting of
      the Association for Computational Linguistics and 8th Conference of the European Chapter of the
      Association for Computational Linguistics (ACL/EACL) (1997).
"Distributional Clustering of English Words." Proceedings 31st Annual Meeting of the Association for
      Computational Linguistics (ACL)(1993). With Fernando Pereira and Naftali Tishby.

Yuying Li
Senior Research Associate
yuying@cs.cornell.edu
http://www.cs.cornell.edu/home/yuying.html
Ph.D. Waterloo University, 1988

My general research interests include numerical optimization and scientific computation. In addition, I am interested in the application of optimization methods to medical, engineering, and financial problems.
My current interests focus on solving nonlinear constrained problems for which the gradient computation is expensive and inexact. These problems arise from many financial application problems.

Lectures
Lung Nodule Segmentation Using Optimization. The Workshop on Mathematics in Image Processing 2000,
      The University of Hong Kong (December 14-16, 2000).
Reconstructing the Unknown Local Volatility Function. A special session of mathematics/optimization for
      finance at Optimization 2001, Aveiro, Portugal (July 23-25, 2001).

Publications
"A Trust Region and Affine Scaling Interior Point Method for Nonconvex Minimization with Linear Inequality
      Constraints." Mathematical Programming Series A, 88(1):1-32 (2000).

Rajit Manohar
Assistant Professor
Member of the School of Electrical and Computer Engineering
and the Graduate Field of Computer Science
rajit@csl.cornell.edu
http://vlsi.cornell.edu/~rajit/
Ph.D. California Institute of Technology, 1998

My research is concerned with the design of efficient asynchronous computation structures in VLSI and the use of formal methods to guarantee the correctness of such structures.
     In work on formal methods, I have developed a new way to analyze the correctness of a class of program transformations commonly used in asynchronous VLSI synthesis. The technique "decompiles" the circuit into a higher level programming language, and provides high-level information about the effect of applying the transformation.
     The amount of power required by a processor is quickly becoming a design constraint. In work on energy-efficient computation, I have developed a new adaptive number representation that significantly reduces the amount of energy required during instruction execution in a general-purpose processor.
     The presence of precise exceptions in a processor can complicate its design. In work with Mika Nystrom and Alain J. Martin, I developed a simple, distributed implementation of an algorithm that implements precise exceptions in asynchronous processors, along with a proof of its correctness using program transformations. An instance of this mechanism was used in the design of an asynchronous MIPS processor.

Awards/Honors
Sonny Yau '72 Excellence in Teaching Award (2001).
IEEE Teacher of the Year Award (2001).
NSF Faculty Early Career Development Award (2000-2004).
Tau Beta Pi and Cornell Society of Engineers Excellence in Teaching Award (2000).

University Activities
Member: Graduate Admissions Committee.
Member: Fields of Electrical Engineering; Computer Science; Applied Math.

Professional Activities
Program Committee : IEEE/ACM Symposium on Advanced Research in Asynchronous Circuits and Systems
      (March, 2001).
Workgroup Organizer: NSF Workshop on Neuromorphic Engineering (July, 2000).

Lectures
Width-adaptive Data Word Architectures. 2001 Conference on Advanced Research in VLSI (March, 2001).
An Analysis of Reshuffled Handshaking Expansions. IEEE/ACM Symposium on Asynchronous Circuits and
      Systems (March, 2001).
Low Energy Adaptive Processors. Computer Science Colloquium, Cornell University, Ithaca NY (October,
      2000).
A Case for Asynchronous Computer Architecture. ISCA Workshop on Complexity-effective Design (June,
      2000).

Publications
"Width-adaptive Data Word Architectures." Proc. 19th Conference on Advanced Research in VLSI,112-129
      (2001).
"Precise Exceptions in Asynchronous Processors." Proc. 19th Conference on Advanced Research in VLSI,16-
      28 (2001). With Mika Nystrom, and Alain J. Martin.
"An Analysis of Reshuffled Handshaking Expansions." Proc. 7th IEEE/ACM Symposium on Asynchronous
      Circuits and Systems, 96-105 (2001).
"A Case for Asynchronous Computer Architecture." Proc. ISCA Workshop on Complexity-effective Design
      (2000).
"A Case for Asynchronous Active Memories." Proc. ISCA Workshop on Solving the Memory Wall (2000). With
      Mark Heinrich.

Greg Morrisett
Associate Professor
jgm@cs.cornell.edu
http://www.cs.cornell.edu/home/jgm/
Ph.D. Carnegie Mellon University, 1995

My research interests are in the design, semantics, and implementation of programming languages. I am particularly interested in exploring how language technology can be used to build reliable, secure, and high-performance systems software. The unifying thread for all my research is the application of advanced semantic constructs in real-world applications.
     Recently, I have concentrated on type systems and logics for enforcing security properties in low-level code. One byproduct of this research is the design of a type system called TAL (Typed Assembly Language) for the Intel x86 architecture and a suite of tools that can efficiently type check binary code. The TAL type system is sufficiently expressive that we can efficiently compile a variety of high-level safe languages, such as ML, Scheme, Safe-C, and Java, to type-correct assembly code. This technology provides a means to safely extend systems such as kernels, web browsers, or hand-helds without the overheads of a virtual machine or just-in-time compiler. Recent work on TAL includes extending the type system to provide enforcement of other security policies, such as resource constraints, that are outside the scope of traditional type systems. I am also studying how other language technologies, such as binary rewriting and inlined reference monitors, can be combined with types and logics to support a wider class of important security properties.
     Another interest is in language, compiler, and runtime support for application-specific memory management. Though standard garbage collection techniques provide a safe and convenient programming model, many systems applications cannot tolerate the overheads introduced by a general-purpose collector. My research here focuses on a range of topics from fast conservative garbage collectors to advanced type systems for region-based memory management, to generalizations of linear type systems. The goal in all of this work is to provide the programmer with more control over memory management without sacrificing safety.
     My other recent interests are in run-time code generation and modal type systems, efficient data representation, and rich forms of polymorphism. Many of these issues are being explored in the context of Cyclone, a next-generation systems language that is being developed jointly between researchers at Cornell and AT&T Laboratories.

Awards
Allen Newell Medal for Research Excellence (2001).
Ralph Watts Excellence in Teaching Award (2001).
Presidential Early Career Award for Scientists and Engineers (2000).
NSF Faculty Early Career Development Award (1999).
Sloan Fellow (1998).

University Activities
Ph.D. admissions committee (1998-2000).
ECE hiring committee (2000).

Professional Activities
Editor: Journal of Functional Programming.
Associate Editor: ACM Transactions on Programming Languages and Systems.
Member: IFIP Working Group 2.8 on Functional Programming.
Participant: INFOSEC Research Council Study Group on Malicious Code; DARPA ISAT Study on Mobile
      Code.
Program Committee: Symposium on Principles of Programming Languages (PoPL 2001); International
      Conference on Functional Programming (ICFP 2000); Principles and Practice of Declarative Programming
      (PPDP 2000); International Symposium on Memory Management (ISMM 2000); Principles of Programming
      Languages (PoPL 2002); Workshop on Semantics, Applications, and Implementation of Program
      Generation (SAIG 2001); Workshop on Multi-language Infrastructure and Interoperability (BABEL 2001).

Lectures
Language-based Security. Danish Technical Institute (ITU), Copenhagen, Denmark (June, 2001).
Towards Next Generation Low-level Languages. University of Minnesota, Minneapolis, MN, (April, 2001);
      Harvard University, Cambridge, MA (February, 2001).
Next-generation Low-level Languages. Workshop on Semantics, Program Analysis, and Computing Environments for Memory Management, London, England (January, 2001).
Mobile Code Security: An Overview. TARA Review, Air Force Research Laboratory (March, 2000).
The Role of Type Systems in Mobile Code Security. DARPA ISAT Study Group on Mobile Code (January,
      2000).
Mobile Code Security. Air Force Scientific Advisory Board (December, 1999).
Advanced Type Systems for Low-level Languages. OpenSIG Conference (October, 1999).
Why Languages and Compilers Matter. INFOSEC Research Council Study Group on Malicious Code (October,
      1999).

Publications
"Syntactic Type Abstraction." Transactions on Programming Languages and Systems 22(6)1037-1080
      (November, 2000). With D. Grossman, and S. Zdancewic.
"Attacking Malicious Code: A Report to the INFOSEC Research Council." IEEE Software 17(5)
      (September/October, 2000) With G. McGraw.
"Alias Types for Recursive Data Structures." In ACM Workshop on Types in Compilation, Montreal, Canada
      (September, 2000). With D. Walker.
"Scalable Certification for Typed Assembly Language." In ACM Workshop on Types in Compilation, Montreal, Canada (September, 2000). With D. Grossman.
"Typed Memory Management via Static Capabilities." Transactions on Programming Languages and Systems
      22(4):701-771 (July, 2000). With D. Walker, and K. Crary.
"Alias Types." European Symposium on Programming, Berlin, Germany (March, 2000). With F. Smith, and D.
      Walker.
"Type Structure for Low-level Programming Languages." 1999 International Colloquium on Automata,
      Languages, and Programming . With K. Crary.
"From System F to Typed Assembly Language." ACM Transactions on Programming Languages and Systems
      21(3):528-569 (May 1999). With D. Walker, K. Crary, and N. Glew.
"Principals in Programming Languages: A Syntactic Proof Technique." In the 1999 International Conference on
      Functional Programming, Paris, France, 197-207 (September, 1999). With S. Zdancewic, and D.
      Grossman.

Andrew Myers
Assistant Professor
andru@cs.cornell.edu
http://www.cs.cornell.edu/andru
Ph.D. Massachusetts Institute of Technology, 1999

Semantic information about programs and data, obtained from the programming language level, provides leverage for addressing difficult problems in computer systems. Programming language ideas can be applied effectively to problems in security, systems, and databases. I am particularly interested in using language-level information to improve security guarantees, performance, and transparency for distributed systems and mobile code.
      One example of this approach is our current work on the problem of protecting confidential data. Current trends are making this problem both more important and more difficult. Computer systems are nearly completely connected via the Internet, allowing software to disseminate private information to almost any location. In addition, we increasingly use untrusted software; for example, downloaded software such as applets. Standard access-control mechanisms are inadequate because they do not control information propagation.
     Static information control is a promising approach for confidentiality. Our programming language Jif judiciously extends Java with privacy annotations that facilitate static analysis of information flows within programs. We are applying the Jif programming model to distributed systems, in which not only programs but also hosts may be untrusted. Computations spanning a network require a protocol for the communicating hosts; we can show through static analysis of such a distributed program when its protocol releases unintended information to untrusted hosts participating in the computation. To develop a theory for the security of such programs in the presence of downgrading channels, we have introduced the idea of 'robust declassification,' which captures the idea that untrusted hosts are unable to exploit these channels.

Professional Activities
Program Committee Member: 2001 IEEE Symposium on Security and Privacy; 18th ACM Symposium on
      Operating Systems Principles (SOSP).

Lectures
Protecting Confidentiality Against Untrusted Programs and Hosts. Mini-Workshop on Mobile Objects/Code and
      Security, University of Tokyo, Japan (October, 2000).
Enforcing Confidentiality in Low-level Programs. SDI/LCS SeminarFest, Carnegie Mellon University, Pittsburgh,
      PA (June, 2001).

Publications
"Robust Declassification." Proceedings of the 14th IEEE Computer Security Foundations Workshop, Cape
      Breton, Nova Scotia, Canada (June, 2001). With Steve Zdancewic.
"Secure Information Flow and CPS." Proceedings of the 10th European Symposium on Programming, Genova,
      Italy (April, 2001). With Steve Zdancewic.

Anil Nerode
Professor
Member of the Department of Mathematics
and the Graduate Field of Computer Science
anil@math.cornell.edu
www.math.cornell.edu/~anil/
Ph.D. University of Chicago, 1956

A decade in the works, the book Automata Theory and its Applications, by Bakhadyr Khoussainov and myself, was published in February, 2001 by Birkhauser (ISBN 0-8176-4027-2). This book makes available in one place material that is used all the time for decidability results in computer science. The basic theme is the correspondence between classes of automata and languages for finite automata, Buchi Automata, Rabin automata and the corresponding games and strategies for those games. It is suitable for a one semester graduate course or a two-semester undergraduate course. "Constructive Concurrent Dynamic Logic," with Duminda Wijesekera, was finished this year and will appear in the Annals of Pure and Applied Logic shortly. This was a project also of 10 years duration, primarily because of the number of cases involved in the intuitionistic treatment, which fortunately decreased from 120 to about 50, for the completeness theorem.
My principal project is a research monograph with engineer Wolf Kohn on the use of Finsler manifolds we associate with distributed optimal control problems throughout engineering to extract close-to-optimal controls in the form of finite automata. We dubbed this area Hybrid Systems in 1991, and many now work in it.
Kohn and I founded a research and development company, Hynomics, some years ago, with venture capital, in Seattle, and it is about to release to a client a prototype 25,000 agent distributed system for supply chain management, entirely based on new mathematical-computer science technology.

Publications
Foreword to: Principles of Modeling and Asynchronous Distributed Simulation of Complex Systems by S.
      Ghosh, IEEE Press (2000).

Keshav Pingali
Professor
pingali@cs.cornell.edu
http://www.cs.cornell.edu/Info/Projects/Bernoulli
Ph.D. Massachusetts Institute of Technology, 1986

My research group works on programming languages and compiler technology for program understanding, restructuring, and optimization. Our goal is to develop the algorithms and tools that are required to raise the level of abstraction at which people program computers, freeing them from having to worry about low-level details of machine architectures, memory hierarchies, etc.
     Our current focus is making programs adapt to hardware faults. This problem has been studied by the operating systems community, but proposed solutions such as message logging have a large run-time overhead. We are developing program analysis and transformation techniques to reduce this run-time overhead by exploiting information about program behavior that can be deduced at compile-time. We plan to deploy these techniques in the Adaptive Software Project, which is a multi-year, multi-institutional project funded by the NSF as part of its Information Technology Research (ITR) initiative.
     We have continued our work on technology for optimizing the performance of programs running on machines with deep memory hierarchies. A recent break-through was the invention of fractal symbolic analysis, which is a powerful program analysis technique that permits compilers to restructure complicated programs far beyond the reach of conventional compilers that use dependence analysis. Traditional symbolic analysis is powerful but it is intractable for most programs. To circumvent this problem, fractal symbolic analysis analyzes a program and its transformed version by repeatedly simplifying these programs until symbolic analysis becomes tractable, ensuring that equality of simplified programs is sufficient to guarantee equality of the original programs. We have shown that this approach is adequate for restructuring codes like LU factorization with pivoting, which have not yielded to previous techniques in the literature. The fractal approach to analysis is likely to prove useful for proving other program properties.
     We are continuing our work on next-generation generic programming techniques. In our approach, algorithm implementors use a different API than data structure designers, and the gap between these APIs is bridged by a compiler. One view of this approach is that it exploits restructuring compiler technology to perform a novel kind of template instantiation. We are demonstrating the usefulness of this new technology by deploying it in a system that generates efficient sparse codes from high-level algorithms and specifications of sparse matrix formats.
     These ongoing projects build on our earlier work on restructuring compilation technology. Our group implemented one of the first compilers that generated code for distributed memory machines, starting from sequential shared memory programs. We introduced techniques called runtime resolution and owner-computes rule, which have now become standard in the area. Our work on linear loop transformations for enhancing parallelism and locality has been incorporated by Hewlett-Packard into its entire compiler product line. We also developed fast algorithms for program analysis problems such as computing the control dependence relation, the static single assignment form of a program, and dataflow analyses. Many of these algorithms have been incorporated into commercial and research compilers.

University Activities
Member: Cornell Theory Center Advisory Committee.

Professional Activities
Program Committee: ACM Symposium on Programming Languages Design and Implementation (PLDI '01);
      International Conference on Supercomputing (ICS '01); EuroPar '01, Supercomputing 2001; HiPC 2001.
Consultant: Grammatech Inc.
Referee/Reviewer: ACM TOPLAS; IEEE Trans. Computers; Journal of Parallel and Distributed Computing;
      Journal of Supercomputing; IEEE Computer; Software Practice and Experience;
Editorial Board: Int. Journal of Parallel Programming; and Discrete Mathematics and Theoretical Computer
      Science.

Lectures
Crash Recovery for Long-running Scientific Applications. IBM T. J. Watson Research Center, Yorktown
      Heights, NY (April, 2001).
Fractal Symbolic Analysis. Indian Institute of Technology, Kanpur (January, 2001).
Automatic Synthesis of Locality Enhancing Transformations for Imperfectly-nested Loop Nests. IBM T.J.
      Watson Research Center, Yorktown Heights, NY (November, 2000).
Synchronization on Blue Gene. IBM T.J. Watson Research Center, Yorktown Heights, NY (October, 2000).

Publications
"Tiling Imperfectly-nested Loop Nests." Supercomputing 2000. With Nawaaz Ahmed, and Nikolay Mateev.
"Automatic Generation of Block-recursive Codes." EUROPAR 2000, 125-134. With N. Ahmed.
"Left-looking to Right-looking and Vice Versa: An Application of Fractal Symbolic Analysis to Linear Algebra
      Code Restructuring." EUROPAR 2000, 155-164. With N. Mateev, and V. Menon.
"A Framework for Sparse Matrix Code Synthesis from High-level Specifications." Supercomputing 2000. With
      Nawaaz Ahmed, Nikolay Mateev, and Paul Stodghill.
"Landing CG on EARTH: A Case Study of Fine-grained Multithreading on an Evolutionary Path." Super-
      computing 2000. With Kevin B. Theobald, Gagan Agrawal, Rishi Kumar, Gerd Heber, Guang R.Gao, and
      Paul Stodghill.

Robbert van Renesse
Senior Research Associate
rvr@cs.cornell.edu
http://www.cs.cornell.edu/home/rvr/
Ph.D. Vrije University, Amsterdam, 1989

I am currently interested in locating and aggregating data in a scalable manner. In the Astrolabe system, we construct a hierarchical distributed database. Only the leaves are writable. The internal nodes are generated at run-time using aggregation dictated by SQL queries that are installed on-the-fly. All internal communication is done using a peer-to-peer epidemic protocol. Astrolabe incorporates an hierarchical public key infrastructure for security.
     My other interests include applying formal methods to systems problems and networking support for peer-to-peer applications.

Professional Activities
Vice-president, Research: Reliable Network Solutions, Inc.
Technical Consultant to FAST.

Publications
"Scalable Fault-tolerant Aggregation in Large Process Groups." International Conference on Dependable
      Systems and Networks, Goteborg, Sweden (July, 2001). With Indranil Gupta, and Kenneth P. Birman.
"Spinglass, Scalable and Secure Communication Tools for Mission-critical Computing." DARPA Information
      Survivability Conference and Exposition II (DISCEX 2001), IEEE Computer Society Press (June, 2001).
      With Ken Birman and Werner Vogels.
"Protocol Switching: Exploiting Meta Properties." International Workshop on Applied Reliable Group
      Communication at the International Conference on Distributed Computing Systems (ICDCS), Phoenix, AZ
      (April, 2001). With Xioming Liu, Mark Bickford, Cristoph Kreitz, and Robert Constable.

Mats Rooth
Professor
FCI, Joint with Linguistics
mr29@cornell.edu
Ph.D. University of Massachusetts, 1995

Over the past several years, in collaboration with several colleagues, I have been developing an approach to research and applications in linguistics and computational linguistics that combines theoretical-linguistic formalisms, knowledge, and problem statements with numerical modeling and parameter estimation techniques. We have developed and implemented a grammatical formalism known as head-lexicalized probabilistic context free grammar and have employed it in large-scale experiments on German and English concerned with learning lexical information from text corpora. The methods have the status of a basic language technology, which can be applied in numerous ways in research and applications. My own interests relate mainly to scientific research in linguistics, particularly syntax, semantics and lexical semantics.
     The work aims to devolop an experimental paradigm for work in linguistics in which lingustic events are observed in large text corpora using a robust parser and used to develop and test theories. Linguistic theory is placed in contact with observations on a massive scale using mixed numerical/symbolic comptutational models, and much relevant knowledge is learned from data using numerical optimization algorithms.
The linguistic representations assumed in this work are those of linguistic theory, for instance grammars, labeled trees and structured lexical entries. In the numerical modeling component, grammars and the lexicon are given a probabilistic interpretation, as defining a parameterized family of probability distributions. Many problems can then be solved using probabilistic optimization algorithms. For instance, inducing a lexicon from a text corpus amounts to maximizing a certain polynomial.
     I also work on the semantics of human languages, using logical and denotational-semantic techniques. Currently, I am investigating connections between the syntax and semantics of elliptical constructions and the semantics and phonology of intonation.

University Activities
Administration of Computational Linguistics Lab.
Graduate Admissions Committee.
Phonetics Search Committee.
Founders' Committee, Faculty of Computing and Information.
Social Sciences Advisory Council.

Professional Activities
2000 Workshop on Language Engineering, Johns Hopkins University. (Five-week workshop in which a reading
      comprehension prototype was implemented).
Editorial board: Natural Language Semantics.

Lectures
Parse Forest Computation of Expected Governors. Colloquium presentation, University of Sussex (January,
      2001).
Inducing a Semantically Annotated Lexicon. Symposium From Signals to Structured Communication, Cornell
      University (May 4 and 5, 2001).
Empty Domain Effects for Presuppositional and Non-presuppositional Determiners. Conference on
      Presupposition, Stuttgart (September, 2000).

Publications
"Parse Forest Computation of Expected Governors." Proceedings of the 39th Annual Meeting of the
      Association for Computational Linguistics (2001). With Helmut Schmid.
"Inducing a Semantically Annotated Lexicon via EM-based Clustering." Proceedings of the 37th Annual
      Meeting of the Association for Computational Linguistics (1999). With Stefan Riezler, Detlef Prescher,
      Glenn Carroll, and Franz Beil.
"Inside-outside Estimation of a Lexicalized PCFG for German." Proceedings of the 37th Annual Meeting of the
      Association for Computational Linguistics (1999). With Franz Beil, Glenn Carroll, Detlef Prescher, and
      Stefan Riezler.
"Valence Induction with a Head-lexicalized PCFG." Proceedings of the 3rd Conference on Empirical Methods
      in Natural Language Processing (1998).
"On the Interface Principles for Intonational Focus." Proceedings of the 6th Conference on Semantics and
      Linguistic Theory (1996). With Glenn Carroll.

Fred B. Schneider
Professor
Director, Information Assurance Institute
fbs@cs.cornell.edu
http://www.cs.cornell.edu/People/fbs/
Ph.D. SUNY Stonybrook, 1978

Research this past year continued into techniques to support the construction of concurrent and distributed systems for high-integrity, mission-critical settings. Security and fault-tolerance are of paramount concern here.
     Inlined reference monitors (IRMs) remain a promising approach for enforcing security policies, especially the fine-grained policies needed when the Principle of Least Privilege is employed to protect against hostile mobile code. This past year, we concentrated on putting the approach into practice. Cornell's Digital Library Project is looking to PSLang/PoET for imposing rights management and preservation policies on digital objects in their repository; we have prototyped a "digital university" to understand the enforcement issues there. And we started investigating the deployment of IRMs in the Windows operating system.
     Our investigations into interactions between security and fault-tolerance (joint work with Lidong Zhou and Robbert van Renesse) also continued. We have deployed COCA, our replicated certification authority, to four sites on the Internet and made detailed performance measurements. The system supports on-line revalidation of name-key bindings and is designed to resist a broad collection of denial of service attacks.

University Activities
Sabbatical leave (2000-2001).
Founders Committee, Faculty of Computing and Information.
Engineering College Teaching Awards Committee.

Professional Activities
Director: AFRL/Cornell Information Assurance Institute.
Editor: Distributed Computing; Information Processing Letters; High Integrity Systems; Annals of Software Engineering; ACM Computing Surveys.
Co-managing Editor: Texts and Monographs in Computer Science, Springer-Verlag.
Chairman: International Review of UK Computer Science Research.
Program committee: NORDSEC 2000 Fifth Nordic Workshop on Secure IT Systems-Encouraging Co
      -operation; 3rd Information Survivability Workshop (ISW-2000); Information/System Survivability Workshop
      2001; 2001 Usenix Security Symposium; 2000 PODC Influential Paper Award.
Industrial Advisory Committees: JavaSoft Security Advisory Committee; deCode Genetics Security Advisory
      Board; Eweb University.Com Board of Advisors; JXTA Technical Advisory Council; CIGITAL Technical
      Advisory Board; Fast Search and Transfer Technical Advisory Board.
Other Advisory Committees: MITRE Corporation, Research and Technology committee of the Board of
      Directors; UK Dependability Interdisciplinary Research Collaboration (DIRC), Steering Committee.
IFIP Working Group 2.3 (Programming Methodology).

Honors
Fellow, American Association for Advancement of Science (1992).
Fellow, Association for Computing Machinary (1994).
Professor-at-large, University of Tromsoe, Tromsoe, Norway (1996-2004).
Daniel M. Lazar Excellence in Teaching Award (2000).

Lectures
The Case for Language Based Security. Invited Lecture. Informatics-10 Years Back, 10 Years Ahead.
     Saarland University, Saarbruken, Germany (August, 2000).
In-lined Reference Monitors. Microsoft Research. Redmond, Washington (October, 2000).
Radio Interview, "All Things Considered" (October 27, 2000).
The Case for Language Based Security. IFIP wg2.3. Santa Cruz, CA (January, 2001).
The Design and Deployment of COCA. Distinguished lecture series. SUNY Stony Brook. Stony Brook, NY
      (February, 2001).
Fast P2P Possibilities. Tysil, Norway (February, 2001).
The Design and Deployment of COCA. Department of Computer Science. University of Tromso. Tromso,
      Norway (February, 2001).
The Case for Language Based Security. Keynote Address, ACM Southeast Conference 2001, Athens, GA
      (March, 2001).
The Design and Deployment of COCA. Department of Computer Science. University of Texas, Austin, TX
      (March, 2001).
The Case for Language Based Security. IBM Corporation Hawthorne, NY (April, 2001).
The Design and Deployment of COCA. AFOSR Principal Investigators Meeting. Ithaca, NY (May, 2001).
     -. Information Assurance Institute Seminar Series. AFRL/IF Rome Research Site, Rome, New York (June,
      2001).
The Case for Language Based Security. AFCEA Conference. Hamilton, NY (June, 2001).
Escaping the Ivy Tower: Transitioning Technology from a University. AFCEA Conference. Hamilton, NY (June,
      2001).

Publications
"Enforceable Security Policies." ACM Transactions on Information and System Security 3(1):30-50 (February,
      2000).
"Formalizations of Substitutions of Equals for Equals." Millennial Perspectives in Computer Science,
      Proceedings of the 1999 Oxford-Microsoft Symposium in honour of Professor Sir Antony Hoare, (Davies,
      Roscoe, and Woodcock, editors) Palgrave Publishers, Hampshire, England, 119-132 (November, 2000).
      With David Gries.
"Editorial: Time for Change." Distributed Computing 13(4):187 (November, 2000).
"A Language-based Approach to Security. " Informatics- 10 Years Back, 10 Years Ahead. Lecture Notes in
      Computer Science 2000 (Reinhard Wilhelm, editor), Springer Verlag, Heidelberg, 86-101, (2000). And Greg
      Morrisett, Robert Harper.

David Schwartz
Assistant Professor
dis@cs.cornell.edu
http://www.cs.cornell.edu/dis/
Ph.D. State University of New York at Buffalo, 1999

My main area of research is in computational mechanics. My recent work focuses on applying mathematical techniques of deterministic uncertainty to enhance methods of structural engineering analysis. I am applying one such technique, called Interval Analysis, which adapts traditional numerical operations by replacing numbers with intervals that model uncertain values. This set-based form of structural analysis uses discrete mathematics to perform types of parametric studies, that traditional techniques cannot. However, the interval approach introduces numerical inaccuracies in solutions that map to infeasible structural behaviors. A major goal of this research involves improving the quality of the interval-based solutions to reduce these inaccuracies to produce a technique suitable for design engineers.
     This year I restructured the introductory programming courses to update the technology and problem-solving tools being taught. For the first time, students were introduced to Maple in CS99, a preliminary course in programming that provides experience to students before they take other courses. MATLAB has also received greater focus in a new introductory programming course, CS100M. It has generated quite a bit of excitement from students and other universities because of the equal focus on both MATLAB and Java to help teach programming using problems of computational science. I have also been coordinating the Computer Science Department's "short courses," which are 1-credit courses taught by Ph.D. students. These courses focus on teaching computer languages and help the student instructors gain teaching experience. The instructors turn to me for advice on curriculum development, teaching methods, and navigating university policies. I have created a new short course, CS214: Advanced UNIX Tools, which will be taught for the first time in the next academic year.
     Continuing my work from spring 2000, colleagues in the College of Engineering and I converted the pilot CS100 Academic Excellence Workshop (AEW) into a full-fledged program that runs in conjunction with CS100M. We continued to pilot AEWs for CS100J, which is the more traditional programming course. The AEW fosters cooperative learning, where students work in teams to solve challenging problems in a non-competitive environment. Students have responded enthusiastically, so we began designing a new lab in spring 2001 to expand our capacity, especially to help the CS100J pilot grow. Since AEWs tend to entice students, especially women and minorities, we will be tracking engineering and computer science enrollments in the next academic year to measure the effectiveness of our AEW program in increasing diversity in the sciences. I look forward to working on the CS AEW with Daisy Fan, whom I recruited into the department.

University Activities
Faculty Advisor: Association of Computer Science Undergraduates.
Coordinator: Academic Excellence Workshop (AEW) for CS100 program.
Coordinator: departmental short-course advice and instruction.
Member: Student Experience Committee.
Member: Computing Policy Committee.

Professional Activities
Textbook Reviewer: McGraw-Hill.

Lectures
The Inexperienced Educator's Guide To Managing A Large Hierarchical Staff. Emerging Technologies for Industry and Education, Annual Meeting of the St. Lawrence Section, American Society Engineering Education (March, 2001).

Bart Selman
Associate Professor
selman@cs.cornell.edu
http://www.cs.cornell.edu/home/selman
Ph.D. University of Toronto, 1991

The focus of my research is on computation intensive methods in artificial intelligence, in particular fast general reasoning, search, and planning techniques. I also investigate the various sources of complexity in hard computational problems. In this work, I explore connections between computer science, artificial intelligence, and statistical physics. In addition, I study issues in problem representation, including the robustness of encodings, abstraction, compilation, and approximation methods. These issues are critical to the successful application of reasoning and search methods in realistic domains. In terms of applications, I consider challenge problems from planning, knowledge representation, multi-agent systems, and machine learning. Our planning system, Black Box, developed jointly with Henry Kautz of the University of Washington, is one of the fastest general purpose planning systems. Finally, in recent projects, I am exploring connections between machine learning methods and reasoning and planning techniques. In joint work with groups at the University of Washington and Microsoft Research, I study the use of Bayesian machine learning techniques for dynamic adaptive control of computational resources.

Awards
Fellow: American Association for Artificial Intelligence.
Alfred P. Sloan Research Fellowship (1999-2001).
NSF Faculty Early Career Development Award (1998-2002).

University Activities
Chair: Bits On Our Minds Science Fair.
Coordinator of the AI Seminar Series.
Cognitive Studies Undergraduate Committee.
Member: Fields of Cognitive Studies and Applied Mathematics.

Professional Activities
Executive Council: American Association for Artificial Intelligence.
Editorial Board: Constraints: An International Journal; Annals of Mathematics; and Artificial Intelligence.
Advisory Board: Journal of Artificial Intelligence Research (JAIR).
Guest Editor: Theoretical Computer Science; Discrete Applied Mathematics; Electronic Notes in Discrete
      Mathematics; and Annals of Mathematics and Artificial Intelligence.
Organizer and program co-chair: SAT-2001: Workshop on Theory and Applications of Satisfiability Testing.
Program committee: AAI-2000 Workshop on Leveraging Probability and Uncertainty in Computation; 17th Intl.
      Conf. on Artificial Intelligence (IJCAI-2001); AAAI-2001 Symposium on Uncertainty in Computation.Referee/Reviewer: Artificial Intelligence Journal (AIJ); JACM Constraints: An International Journal; Journal of Artificial Intelligence Research (JAIR); NSF Journal of Automated Reasoning; Theoretical Computer Science; Science.

Lectures
Insights from Statistical Physics into Computational Complexity. Colloquium, California Institute of Technology
      (Caltech), Pasadena, CA (October, 2000).
Survey of Artificial Intelligence and Knowledge Representation. Six lectures at AFRL/IF, Rome, NY
      (October/November, 2000).
Understanding Complexity: Recent Developments and Directions. Colloquium, University of Minnesota,
      Computer Science, Minneapolis, MN (October, 2000).
Principled Analysis and Synthesis of Agent Systems. Meeting on Taskable Agent Software Kit, DARPA,
      Charlotte, NC (October, 2000).
Challenge Problems for Propositional Reasoning and Search. Distinguished Lecture Series, 7th Annual
      Knowledge Representation and Reasoning Lecture, York University, York, UK (November, 2000).
Understanding Complexity: Recent Developments and Directions. Colloquium, Leeds University, Leeds, UK
      (November, 2000).
Controlling Complexity: Structured Problems and Backbone Variables. Meeting on Autonomous Negotiation
      Teams, DARPA, Charlotte, NC (November, 2000).
Satisfiability Testing: Recent Developments and Challenge Problems Plenary Lecture, AAAI Spring
       Symposium on Answer Set Programming, Palo Alto, CA (March, 2001).

Publications
"A Bayesian Approach to Tackling Hard Computational Problems." Proc. 17th Conf. on Uncertainty and
      Artificial Intelligence (UAI-2001), Seattle, WA (2001). With E. Horvitz, Y. Ruan, C. Gomes, H. Kautz, and
      M. Chickering.
"Balance and Filtering in Structured Satisfiable Problems." Proc. 17th Intl. Conf. on Artificial Intelligence
      ( IJCAI-2001), Seattle, WA (2001). With H. Kautz, Y. Ruan, D. Achlioptas, C. Gomes, and M. Stickel.
"Algorithm Portfolios." Artificial Intelligence Journal 126 (2001). With C. Gomes.
"Distribute Constraint Satisfaction in a Wireless Sensor Tracking System." Proc. Workshop on Distributed
Constraint Reasoning (CONS-2), IJCAI-2001, Seattle (2001). With R. Bejar, C. Gomes, and B. Krishnamachari.
"Analysis of Random Noise and Random Walk Algorithms." Principles and Practice of Constraint Programming (CP 2000). Lecture Notes in Computer Science 894:278-290 (2000). With B. Krishnamachari, Xi Xie, and S. Wicker.
"Compute Intensive Methods in Artificial Intelligence." Annals of Mathematics and Artificial Intelligence 28(1) (2000).
"Heavy-tailed Phenomena in Satisfiability and Constraint Satisfaction Problems." Journal of Automated Reasoning 24(1/2):67-100 (2000). With C. Gomes, N. Crato, and Henry Kautz.

David Shmoys
Professor and Member of the School of Operations Research
and the Graduate Field of Computer Science
shmoys@cs.cornell.edu
http://www.cs.cornell.edu/home/shmoys/shmoys.html
Ph.D. University of California, Berkeley, 1984

The primary focus of my research is on the design and analysis of efficient algorithms for discrete optimization problems, and in particular, on approximation algorithms for NP-hard problems.
     Computational complexity theory provides a mathematical foundation for the intractability of many computational problems by proving that all NP-complete problems are equally hard. However, in practice, real-world inputs for some NP-hard optimization problems are straightforward to solve, whereas for others, even quite modestly sized inputs are beyond the limits of the most sophisticated methods. Analogously, from a theoretical perspective, for some NP-hard optimization problems it is possible to efficiently compute solutions that are guaranteed to be arbitrarily close to optimal, whereas for others, computing even a crude approximation to the optimum is also NP-hard. In fact, the extent to which such approximation algorithms exist for a problem provides a surprisingly accurate theoretical yardstick for its actual computational difficulty.
     Our work has been motivated by the fact that certain linear programming relaxations have been shown to provide extremely good lower bounds on typical data. We provided a theoretical understanding of the strength of these bounds by designing algorithms that "round" the fractional solutions to these linear programs to nearby integer solutions without degrading the quality of the solution too much. We have been investigating a variety of clustering problems, and in joint work with Moses Charikar, Sudipto Guha, and Eva Tardos, we have obtained the first constant factor approximation algorithm for the k-median problem. We have also obtained improved approximation algorithms for the uncapacitated facility location problem (in joint work with Fabian Chudak) and for a variety of more general multi-level facility location problems (in joint work with Karen Aardal and F. Chudak, and with Nathan Edwards). For the former problem, not only is the theoretical performance of our algorithm surprisingly good, but it is more effective in practice than all previously known heuristic procedures for this problem. We have also been investigating the application of some of these rounding methods to problems arising in computational genomics, in joint work with Dan Brown and the group in plant sciences at Cornell, working under Steve Tanksley. The resulting software for computing genetic linkage maps has already seen widespread application within the genomics community.

University Activities
Member: Search Committee for Dean of the Graduate School.
Member: Genomics Task Force.
Member: FCI Working Group on Computational Biology.
Member: FCI Working Group on Computational Science.

Professional Activities
Editor-in-chief: SIAM Journal on Discrete Mathematics.
Editor: SIAM Journal on Computing.
Co-editor: SIAM/MPS Series on Optimization.
Associate Editor: Mathematics of Operations Research; Mathematical Programming; Journal of Scheduling.
Guest Editor: J. `of Algorithms (Special issue devoted to selected papers of the 11th Annual Symposium on Discrete Algorithms).

Lectures
Approximation Algorithms for Facility Location Problems.
-. Invited plenary lecture. CO 2000, Greenwich (July, 2000).
-. Invited plenary lecture. CONF 2000, Saarbrucken, Germany, September 2000.
A Constant-factor Approximation Algorithm for the K-median Problem. International Symposium on Mathematical Programming, Atlanta, GA (August, 2000).
Selective Mapping: A Strategy for Optimizing the Construction of High-density Linkage Maps.
-. Tri-Institutional Workshop on Computational Biology, Cornell ( September, 2000).
-. NIH Symposium - From Genes to Proteins and to Biological Function: Computational Approaches, Cornell (October, 2000).

Publications
"Selective Mapping: A Strategy for Optimizing the Construction of High-density Linkage Maps." Genetics      155:407-420 (2000). With T. J. Vision, D. G. Brown, R. T. Durrett, and S. D. Tanksley.
"Approximation Algorithms for Facility Location Problems." In: Approximation Algorithms for Combinatorial Optimization, K. Jansen and S. Khuller, editors, APPROX 2000, Lecture Notes in Computer Science 1913:27-33, Springer, Berlin (2000).

Landmark Publications
"Using Dual Approximation Algorithms for Scheduling Problems: Theoretical and Practical Results." J. Assoc. Comput. Mach. 34:144-162 (1987). With D.S. Hochbaum.
"An Approximation Algorithm for the Generalized Assignment Problem." Math. Programming 62:461-474 (1993). With E. Tardos.
"Approximation Algorithms for Fractional Packing and Covering Problems." Math. Oper. Res. 20, 257-301 (1995). With S.A. Plotkin and E. Tardos.
"Scheduling to Minimize the Average Completion Time: On-line and Off-line Approximation Algorithms." Math. Oper. Res. 22:513-544 (1997). With L. A. Hall, A. S. Schulz, and J. Wein.
"A Constant-factor Approximation Algorithm for the K-median Problem." Proceedings of the 31st Annual ACM Symposium on Theory of Computing, 1-10 (1999). With M. Charikar, S. Guha, and E. Tardos.

E. Gun Sirer
Assistant Professor
egs@cornell.edu
http://www.cs.cornell.edu/People/egs/
Ph.D. University of Washington, 2001

I am broadly interested in the design and implementation of distributed operating systems for modern networks.
     My main focus is on operating system support for ad-hoc mobile networks. Ad-hoc networking is a newly emerging computing paradigm where disparate, mobile hosts form temporary alliances to accomplish certain tasks. They naturally arise in many real-world settings including disaster relief operations, smart vehicles, and distributed sensor networks. Such ad-hoc settings, characterized by constant change and limited resources, require a high degree of adaptability and resource awareness from applications. Current state of the art, however, requires all such functionality to be manually encoded by the application programmer.
     We are currently building an operating system, called MagnetOS, that provides automatic and transparent distribution of ad-hoc networking applications. MagnetOS makes an entire ad-hoc network appear as a single Java virtual machine to applications. By transparently migrating application components from node to node, MagnetOS can increase system longevity, reduce application delays, and decrease bandwidth consumption, even in the presence of conflicting applications. Consequently, MagnetOS simplifies application design, increases application portability by obviating error-prone manual coding, and achieves an efficient placement of application components within the network that adapt to changes in resource availability and connectivity.
     In conjunction with MagnetOS, we are examining the design of secure, energy efficient operating systems for small, embedded computers. Newly emerging smart cards, which are capable of downloading and executing arbitrary code, are becoming part of our critical infrastructure as they are being deployed in financial and telecommunications networks. We are looking at different strategies for testing their security, as well as evaluating code verification techniques for such severely resource-constrained systems.
     A third research direction looks at distributed systems at the opposite end of the spectrum; namely, the high-performance clusters that make up web sites. We are developing techniques for formally specifying and reasoning about the interfaces they expose to the rest of the network, as well as techniques for enforcing a security policy uniformly across a web site.

Lectures
Distributed Virtual Machines: A New System Architecture for Networked Computers (March-May, 2000).
A New System Architecture for Ubiquitous Computing. EGSA Graduate Engineering Social Seminar Series (March, 2001).

Publications
"Comprehensive Synchronization Elimination for Java." Science of Computer Programming, special issue on
     Static Analysis (April, 2001). With Jonathan Aldrich, Craig Chambers, and Susan Eggers.
"Design and Implementation of a Distributed Virtual Machine for Networked Computers. Proceedings of the
     Seventeenth Symposium on Operating Systems Principles, 202-216, Kiawah Island, SC (December, 1999).
With Robert Grimm, Arthur J. Gregory, and Brian N. Bershad.
"Extensibility, Safety and Performance in the SPIN Operating System. Proceedings of the Fifteenth
      Symposium on Operating Systems Principles, 267-284, Copper Mountain, CO (December, 1995). With
      Brian N. Bershad, Stefan Savage, Przemyslaw Pardyak, Marc Fiuczynski, David Becker, Craig Chambers,
      and Susan Eggers.

Evan Speight
Assistant Professor
Member of the School of Electrical and Computer Engineering
and the Graduate Field of Computer Science
espeight@csl.cornell.edu
http://www.ds-csl.cornell.edu/~espeight/
Ph.D. Rice University, 1998

My research centers around different aspects of distributed computing. In particular, my three areas of focus are software runtime systems for distributed computing platforms, improving the performance of commodity cluster-based shared memory systems, and providing location-independent data access through the concept of affinity directed mobility.
      My work in software distributed system runtime development has resulted in the dissemination of the Brazos system to over 100 registered users worldwide. Brazos is a high performance parallel programming environment distinguished by its use of multithreading, selective multicast, a software-only implementation of scope consistency, and several adaptive runtime performance tuning mechanisms. Brazos supports both shared memory and message passing programming styles, and provides very efficient mechanisms for thread migration and checkpoint/recovery. Currently, my research on parallel programming runtime systems includes the development a version of the Message Passing Interface (MPI) that provides thread migration between nodes in a cluster for load balancing, fault tolerance, and higher performance.
     The emergence of ubiquitous communication infrastructure and high performance, low power computing resources challenges us to explore a better alternative to the current fragmentation of data, applications, and devices that many users are faced with today. The Bifrost location independent computing project seeks to provide a flexible and comprehensive information access environment. The function of Bifrost is to provide location and device independent access to data. Data in Bifrost encompasses both information and the applications used to manipulate that information. Bifrost uses affinity between data, and between users and data, to make decisions about when and where to move data. We refer to this approach as affinity directed mobility.
     The core research issues of the project are mobility management (how we move data and threads to support user and device mobility), data management (how we represent, access, update, and protect information), and application management (how we provide system-wide access to application data). In contrast to previous approaches to mobile data management that employ hoarding in anticipation of disconnection, Bifrost anticipates "connection elsewhere" instead of disconnection. Our rationale for this "almost always connected" approach is as follows: We project that in 3-5 years, the situation will exist in which users will be able to be in an "always connected" state, connected almost anytime and anywhere that they choose. Wireless access points (e.g., 802.11, infrared, and Bluetooth) will be common, even ubiquitous, in nearly every home and public place, as they are today in many situations including coffee shops, airports, and office buildings. Additionally, services such as Infared and Bluetooth will allow connection to the Internet in nearly any setting from almost anywhere. In this situation, the problem of how a user can have unilateral access to any piece of remote personal data, regardless of device or location, becomes increasingly important. Thus Bifrost seeks to provide a common data access model, and exploring the design space while providing a rich set of tools for the manipulation and sharing of personal data is a core focus of the proposed research.

University or Professional Activities
Member: ECE Curriculum and Standard Committee.
Member: Faculty Advisory Board on Information Technologies.
Reviewer: Transactions of Parallel and Distributed Computing.

Lectures
WSDLite: A Lightweight Alternative to Windows Sockets Direct Path. Fourth Windows Symposium, Seattle, WA (August, 2000).
Efficient Parallel Computing on Multiprocessor Clusters. EE Colloquium, Cornell University (March, 2000).

Publications
"WSDLite: A Lightweight Alternative to Windows Sockets Direct Path." Proceedings of the 4th Usenix
      Windows Symposium (August, 2000). With John Bennett, and Hazim Abdel-Shafi.
"Efficient User-level Thread Migration and Checkpointing on Windows NT Clusters." In Proceedings of the 3rd
      Usenix Windows NT Symposium (July, 1999). With Hazim Abdel-Shafi and John Bennett.
"Realizing the Performance Potential of the Virtual Interface Architecture." Proceedings of the International
      Conference on Super computing (ICS) (June, 1999). With John Bennett.

Paul Stodghill
Research Associate
stodghil@cs.cornell.edu
http://www.cs.cornell.edu/stodghil/
Ph.D. Cornell University, 1997

My research interest is in program transformations and program synthesis for computational science applications.
     Increasingly, the frontiers of science are being explored using computer simulation and modeling. Creating these scientific applications poses a number of difficulties for computational scientists. First, these applications are often very complex and may contain hundreds of thousands, and even millions of lines of code. Furthermore, these applications are intended to run on the most advanced, high performance supercomputers. I am interested in developing software that helps to reduce the development cost of these applications. One way to do this is to raise the level of abstraction of programming languages and software libraries in order to bring them closer to the programmer's natural domain of discourse. In order to do this, specialized compilers and other program transformation systems are required in order to transform these abstractions into efficient executable programs.
     An example of this is the work that I have done on sparse compilers with Keshav Pingali and his students. Matrix computations, such as those that perform linear algebra operations, are very naturally expressed using Fortran or Matlab style loops and arrays. However, many matrices that arise in scientific applications are sparse, which means that they consist almost entirely (>99%) of zero values. Modifying matrix algorithms to use special-purpose sparse matrix data structures is a complicated and error-prone process. We have developed software that allows the programmer to express their matrix computations and sparse matrix data structures separately and in their most natural representation. Then, we use program transformations to combine the algorithms and data structures into a single, efficient executable.
     Another area in which I am working is that of fault-tolerance for scientific applications. Very few scientific applications would be considered "mission-critical," but many do run for very long periods of time (e.g., for weeks or months) or in environments in which failures are likely to occur (e.g., the desktop workstations within a department). We are currently investigating a number of different approaches for providing fault-tolerance for scientific applications, but one thing is already clear: different applications running on different computers call for different fault-tolerance solutions. A programmer would certainly like to have their application run in many different computing environments, but interfacing with a number of fault-tolerant systems is not usually realistic. In order to provide a single interface from the programmer's point of view, we are investigating a number of implementation techniques involving software library and program transformation technology. This work is being done as part of a project on Adaptive Software with researchers from other departments here at Cornell, Mississippi State University, the College of William and Mary, and other institutions.
     Apart from my core research agenda, I have worked closely with computational scientists in order to develop novel, high-performance, scientific applications. Apart from being interesting in its own right, this work has given me insight into scientific programming that I have found extremely valuable in my research on program transformations. The most important project that I have been involved with was the Crack Propagation for Teraflop Computers (CPTC) project, whose goal was to develop fracture mechanics software that incorporated a number of recent advances in numerical analysis and computational geometry and which delivered very high performance on teraflop-scale computers. Other researches who worked on this project include, Keshav Pingali, Steve Vavasis, Paul Chew, the Cornell FractureGroup headed by Tony Ingraffea (Civil Engineering), Gao Guong-Rong (ECE, University of Delaware) and Nikos Chrisochoides (CS, College of William and Mary).

Lectures
Next-generation Generic Programming and its Application to Sparse Matrix Computations. ACM International
      Conference on Supercomputing, Sante Fe, NM (May, 2000).
Parallel FEM Simulation of Crack Propagation on the AC3 Velocity Cluster. ACM Second Workshop on
      Cluster Cluster-Based Computing, Sante Fe, NM (May, 2000).
Crack Propagation on Teraflop Computers. AC3 Meeting, Cornell University (June, 2000).

Publications
"A Framework for Sparse Matrix Code Synthesis from High-level Specifications." In Supercomputing 2000,
      Dallas, TX (November 4-11, 2000). With N. Ahmed, N. Mateev, and K. Pingali.
"Landing CG on EARTH: A Case Study of Fine-gained Multithreading on an Evolutionary Path." In Supercomputing 2000, Dallas, TX (November 4-11, 2000). With K. Theobald, G. Agrawal, R. Kumar, G. Heber,
      G. Gao, and K. Pingali.
"Next-generation Generic Programming and its Application to Sparse Matrix Computations." In International
      Conference on Supercomputing, 2000. With N. Mateev, K. Pingali, and V. Kotlyar.
"Parallel FEM Simulation of Crack Propagation on the AC3 Velocity Cluster." In The Second Workshop on
      Cluster Cluster-Based Computing, 2000. With G. Coulouris, G. Heber, D. Lifka, K. Pingali , D. Schneider,
      P. Wawrzynek, and J. Zollweg.
"Parallel FEM Simulation of Crack Propagation - Challenges, Status, and Perspectives." Irregular 2000. With
      B. Carter, et al.

Eva Tardos
Professor
eva@cs.cornell.edu
http://www.cs.cornell.edu/home/eva/eva.html
Ph.D. Eotvos University, Hungary, 1984

My research interest focuses on the design and analysis of efficient methods for combinatorial optimization problems and their applications to various fields. I am mostly working on problems that involve graphs or networks.
     One general area of my research is designing fast algorithms that provide provably close-to-optimal results for NP-hard problems. Although research on polynomial time approximation algorithms started in the1970s soon after the discovery of NP-completeness, it has truly blossomed only in the past decade. Amazing progress has occurred both in our ability to design approximation algorithms, and in proving limits to approximability. Over the last years I have been working on different approximation algorithms on various basic combinatorial problems. I have worked on algorithms for various cut problems, and clustering type problems. These problems are motivated by applications arising in vision, networking and clustering.
     I am also working on the interface of algorithms and game theory. Approximation algorithms provide a tool for understanding issues in game theory, such as evaluating and designing multi-agent games. Such games underlie many phenomena in networking. In joint work with Tim Roughgarden, I worked on understanding the quality of routing obtained if separate agents each make selfish routing decisions in a global network. Our goal is to understand the tradeoffs between introducing a global control mechanism, and the loss in quality obtained by letting each agent make selfish decisions. We use a model where each link in a network has a delay that is an arbitrary monotone and continuous function of the amount of flow on the link. We compare the quality of such a Nash equilibrium with a globally planned optimal routing whose goal is to minimize the sum of all delays. In the case of linear delay functions, we show that the global optimum can be at most a factor of four-thirds better than the Nash equilibrium. For more general delay functions, the value of the global optimum can be arbitrarily better than the Nash equilibrium even when the delay functions are rather simple. On the other hand, we show that for any monotone and continuous delay function the cost of a Nash equilibrium is at most as much as the optimal for the case in which each agent has to send twice as much flow. Roughly speaking, this shows that one can eliminate the need for introducing a global control mechanism at the cost of designing a network that can support twice as much flow.

Awards
Fellow: American Academy of Arts and Sciences (2001).
Fellow: ACM (1998).
Faculty of the Year: Association of Computer Science Undergraduates.

University Activities
Director of Graduate Studies: Computer Science Department, Cornell University
Member: Fields of Operations Research and Applied Mathematics
Member: FACTA, Faculty Advisory Committee on Tenure and Appointment.
Member: WISE, Advisory group on Women in Science and Engineering.

Professional Activities
DIMACS External Advisory Board member, since 1996.
Co-organizer: DIMACS special year on Computational Intractability in1999-2001.
Area editor for Discrete Optimization: Mathematics of Operations Research.
Editor: SIAM Journal of Computer Science; Chicago Journal of Theoretical Computer Science; Combinatorica; Journal of Interconnection Networks.
Program committee: International Workshop on Randomization and Approximation Techniques in Computer
      Science (APPROX), 2000; ACM-SIAM Symposium on Discrete Algorithms (SODA) 2001; ACM
      Symposium on the Theory of Computing (STOC) 2001.

Lectures
A Classification Problem Related to Multi-way Cuts. International Symposium on Mathematical Programming,
      Atlanta GA (August, 2000).
How Bad is Selfish Routing?
-. International Symposium on Mathematical Programming, Atlanta GA (August, 2000).
-. Cornell University, Department of Mathematics, VIGRE Interdisciplinary Colloquium (September, 2001).
-. Cornell University, Computer Science Department, Distinguished Lecture Series (October, 2000).
-. 41st Annual IEEE Symposium on the Foundations of Computer Science, Redondo Beach, CA (November, 
      2000).
-. University of Southern California, Computer Science Department, Distinguished Lecture Series (November, 2000).
Flow-based Algorithms for Some Metric Labeling Problems. International Symposium on Mathematical
      Programming, Atlanta GA (August, 2000).
A Constant-factor Approximation Algorithm for the K-median Problem. International Symposium on
      Mathematical Programming, Atlanta GA (August, 2000).
Classification with Pair-wise Relationships (invited talk). Horizons in Combinatorics, A Conference on Graph
      Theory, Vanderbilt, TN (May, 2001).
Sequence of four lectures on Approximation Algorithms. DONET summer school on Integer and Combinatorial
      Optimization, Utrecht, Netherlands (June, 2001).

Publications
"How Bad is Selfish Routing?" Proceedings of the 41st Annual IEEE Symposium on the Foundations of
      Computer Science (October, 2000). With Tim Roughgarden.
"Allocating Bandwidth for Bursty Connection." SIAM Journal on Computing 30(1):191-217 (February, 2001).
      With Jon Kleinberg, and Yuval Rabani.

Landmark Publications
"Approximation Algorithms for Classification Problems with Pair-wise Relationships: Metric Partitioning and
      Markov Random Fields." In the Proceedings of the 40th Annual IEEE Symposium on the Foundations of
      Computer Science, 14-23 (November, 1999). With Jon Kleinberg.
"A Constant-factor Approximation Algorithm for the K-median Problem. In the Proceedings of the 31st Annual
      ACM Symposium on the Theory of Computing, 1-10 (May, 1999). With Moses Charikar, Sudipto Guha,
      and David Shmoys.
"The Quickest Transshipment Problem." Mathematics of Operations Research 36-62 (February, 2000). With
      Bruce Hoppe.
Fast Approximation Algorithms for Multicommodity Flow Problems. Journal of Computer and System Sciences
      50:228-243 (1995). With Tom Leighton, Fillia Makedon, Serge Plotkin, Cliff Stein, and Spyros Tragoudas.
"A Strongly Polynomial Algorithm for the Minimum Cost Circulation Problem." Combinatorica 5:247-255
      (1985).

Tim Teitelbaum
Associate Professor
tt@cs.cornell.edu
http://www.cs.cornell.edu/Info/People/tt/Tim_Teitelbaum.html
Ph.D. Carnegie Mellon University, 1975

My research is concerned with the use of fine-grain dependence graphs for specification, development, and analysis of software and hardware systems. The objective is a new generation of tools that provide precise and complete information about the structure of complex systems. I am working to improve the performance and functionality of generic dependence-graph technology, and I am also exploring the use of the technology in various application domains.
     Dependence-graph technology can be used in a program understanding system, where the graphs may include forward and backward links between each assignment statement and possible uses of the values stored by that assignment. Pointer analysis can be used so that indirect loads and stores through pointers are taken into account, as well as indirect function calls. Dataflow analysis can be used so that links between unrelated assignments and uses are excluded. Operations that highlight forward and backward slices show the impact of a given statement on the rest of the program (forward slicing) and the impact of the rest of a program on a given statement (backward slicing). Operations that highlight paths between nodes in the dependence graph (chops) show ways in which the program points are interdependent (or independent).
Uses of slicing and chopping include software development, maintenance and re-engineering of legacy code, test-data generation, security-assurance and safety-assurance inspection, and semantic interference checking in configuration management systems.
     I am working with Ph.D. student Lyn Millett, who is studying program dependence-graphs and slicing of concurrent programs.

[On leave spring 2001.]

Professional Activities
Co-founder and Chairman: GrammaTech, Inc.
Member: Science and Technology Study Group, Infosec Research Council (October 1999-May 2000).

Lectures
Static-semantic analysis based on dependence graphs. SPAWAR, San Diego, CA (August, 1999).
-. Hewlett-Packard, Rancho Bernardo, CA (October, 1999).

Publications
"Issues in Slicing Promela and its Applications to Model Checking, Protocol Understanding, and Simulation."
      International Journal on Software Tools for Technology Transfer 2(4):343-349 (2000). With L. Millett.
"Program Slicing of Hardware Description Languages." In 10th IFIP WG10.5 Advanced Research Working
      Conference on Correct Hardware Design and Verification Methods (CHARM '99), Bad Herrenalb, Germany
      (September, 1999). With E.M. Clarke, M. Fujita, P.S. Rajan, T. Reps, and S. Shankar.
"A Case for Channel Analysis: Slicing Promela." Proceedings of the International Symposium on Software
      Engineering for Parallel and Distributed Systems, Los Angeles, CA, 52-61 (May, 1999). With L. Millett.

Herbert Van de Sompel
Assistant Professor
herbertv@cs.cornell.edu
http://www.cs.cornell.edu/people/herbertv/
Ph.D. Ghent University, 2000

During the academic year 2000/2001, I focused on furthering my work in two areas of digital library research that I initiated in 1999: the Open Archives Initiative and open reference linking (OpenURL).
     With the Open Archives Initiative, important progress was made through a generalization of the metadata harvesting specifications that were released as the Santa Fe Convention in early 2000. Those specifications were only applicable to preprint-related metadata, but nevertheless attracted the interest from communities outside of the preprint realm. It was decided to revise the specification in depth, in order to extend its applicability to metadata in general. The revision process took 5 months. In that timeframe, a meeting was organized to discuss issues involved in the generalization; several versions of a new protocol document were compiled; an alpha-testing group consisting of 14 international parties was assembled with the aim of testing and refining those versions. In January 2001, the Metadata Harvesting Protocol of the Open Archives Initiative was publicly released at a meeting in Washington DC. Carl Lagoze (Cornell University CS) and I were appointed Executives of the Open Archives Initiative, overseeing and coordinating the technical activities of the Initiative. The Digital Library Federation and the Coalition for Networked Information provide support for those activities. The Metadata Harvesting protocol attracts broad international attention, and several funding agencies explicitly or implicitly call for proposals that build on the protocol.
     Important progress was also made in the realm of the open reference-linking track of my work. With Patrick Hochstenbach (Ghent University) and Oren Beit-Arie (Ex Libris USA), I had publicly released a draft OpenURL specification in January 2000. The OpenURL specification enables open reference linking by:
Allowing one to reference a scholarly object-as well as elements that describe the context in which the reference is provided-in a consistent way as name=value pairs on an HTTP GET or POST URL.
Allowing for the transfer of that reference-as well as of the contextual elements-to a linking server.
     In December 2001, I filed for standardization of the OpenURL with NISO. Soon thereafter, the standardization request was granted, and a formal standardization committee was formed. Meanwhile, the draft OpenURL specification has already gained broad acceptance in the scholarly information industry, with leading companies such as ISI, EBSCO, Swets, SilverPlatter, etc. supporting it in their production systems. Also, an important OpenURL-based prototype was set up with DOI/CrossRef to demonstrate the feasibility of open reference linking based on DOIs.

Professional Activities
Member: Steering Committee, Open Archives Initiative.
Executive: Open Archives Initiative.
Technical Committee: Open Archives Initiative.
Research Advisory Board: OCLC.
NISO Committee AX (OpenURL standardization)
Co-organizer: Open Archives Initiative Meeting, Washington DC (January, 2001).
Workshop on the Open Archives Initiative and peer-review journals in Europe, Geneva (March, 2001).

Lectures
Interconnecting Distributed Scholarly Information Resources in a Context Sensitive Manner. Invited half-day
      workshop, Access 2000 conference, St. John's, Newfoundland, Canada (September, 2000).
The SFX Framework, the OpenURL and the Open Archives Initiative. Invited presentation, SLA Global 2000
      conference, Brighton UK (October, 2000).
Invited presentation in the session "Digital Libraries and their Role in Knowledge Dissemination and Creation."
       ASIS 2000 Conference, Chicago IL (November, 2000).
Invited presentation in the session "Electronic Pre-print Initiatives: A Discussion on Comparative, Historical and
      Emerging Trends". At the ASIS 2000 Conference, Chicago IL (November, 2000).
The Roof is on Fire. Closing keynote at the fall 2000 meeting of the Coalition for Networked Information. San
      Antonio, TX (December, 2000).
The Open Archives Initiative. Keynote at the Workshop on the Open Archives Initiative and peer-review journals
      in Europe, Geneva (March, 2001).
The Open Archives Initiative and Scholarly Communication. Invited presentation, IATUL Conference, Delft (May,
      2001).

Publications
The Open Archives Protocol for Metadata Harvesting, Herbert Van de Sompel and Carl Lagoze, editors
      (January, 2001). http://www.openarchives.org/OAI/openarchivesprotocol.htm.
"The Open Archives Initiative: Building a Low-barrier Interoperability Framework." JCDL2001 (2001). With C.
      Lagoze. http://www.cs.cornell.edu/lagoze/papers/oai-jcdl.pdf.
"Open Linking in the Scholarly Information Environment using the OpenURL Framework." D-Lib Magazine
      7(3)(March, 2001). With Oren Beit-Arie.      http://www.dlib.org/dlib/march01/vandesompel/03vandesompel.htm.
"Generalizing the OpenURL Framework beyond References to Scholarly Works." D-Lib Magazine
      7(7/8)(July/August, 2001). With Oren Beit-Arie.

Charles Van Loan
Professor and Chair
Department of Computer Science
cv@cs.cornell.edu
http://ww.cs.cornell.edu/cv/
Ph.D. University of Michigan, Ann Arbor, 1973

I continue to work in the computational multilinear algebra area. This includes factorization approaches to various fast transforms, Kronecker product preconditioners, and Kronecker-constrained least squares problems.
     This past year Adam Florence and I perfected our implementation of the fast Gauss transform (FGT) of Greengard and Strain. We also developed effective methods for total least squares and weighted least squares when the data matrix is a Kronecker product.
     Applied Mathematics student Carla Martin and I completed work on a solver for linear systems of the form (A-qI)x = b where A is the product of upper triangular matrices. Our method is an order of magnitude faster than the best of previous techniques.

University of Professional Activities
Chair: Department of Computer Science.
Director of Undergraduate Studies: Department of Computer Science.
Member: FCI Founders.
Member: Core Curriculum Governing Board (Engineering).

Lectures
The Ubiquitous Kronecker Product. Invited Lecture, SIAM Linear Algebra Meeting, Raleigh NC (October, 2000).

Publications
"The Ubiquitous Kronecker Product." Journal of Computational and Applied Mathematics 123:85-100 (2000).
"GEMM-based Level 3 BLAS: Algorithms for the Model Implementations." ACM Transactions on Mathematical
      Software 24:268-302 (1999). With P. Ling and B. Kagstrom.
Computational Frameworks for the Fast Fourier Transform, 273pp., SIAM Publications, Philadelphia, PA.
      (1992).
Matrix Computations, 3d ed., 694pp, Johns Hopkins University Press, Baltimore, MD (1996). With G.H. Golub.
Introduction to Scientific Computation: A Matrix-vector Approach Using Matlab, 2d Ed., 365pp, Prentice-Hall,
      Upper Saddle River, NJ (1999).

Steve Vavasis
Associate Professor
vavasis@cs.cornell.edu
http://www.cs.cornell.edu/home/vavasis
Ph.D. Stanford University, 1989

As computer hardware becomes more powerful, there is a corresponding growth in the demand for more efficient algorithms to solve large-scale scientific problems. My research is on the design and analysis of such algorithms. Two Ph.D. students, V. Howle and G. Jonsson (both of the Center for Applied Mathematics), completed their Ph.D.s working with me during the past year. Howle's thesis was on algorithms for modeling and simulation of AC electric power networks. Utility companies are interested in modeling the behavior of the network in the presence of a fault (closed circuit breaker). The governing equations, called the "swing equations," are nonlinear differential algebraic equations for the rotor angles of the generators in the system. We developed new, more accurate algorithms for the swing equations by solving a linear algebraic subproblem (complex-weighted least squares) more accurately. Work on geometry in scientific computing continues. Jonsson's thesis considers the problem of robust intersection of parametric patches with rays and planes. This problem arises in geometric modeling and mesh generation. Our results show that the problem can be solved accurately by transforming it to a generalized eigenvalue computation using the theory of resultants and other algebraic techniques.
     Four new students have begun to work with me on other problems in scientific computing. We are looking at the following topics: the effect of domain shape on solution to boundary value problems, nonlinear diffusion equations, mesh generation for moving objects such as a heart, and the protein model-fitting problem of x-ray crystallography.

University Activities
Member: Graduate admissions committee, Applied Mathematics.
Graduate admissions committee, Computer Science.
Faculty Senate.
University Hearing Board.
BOOM 2001 Vice-chair.

Professional Activities
Editorial Boards: Journal of Global Optimization; SIAM Journal Matrix Analysis and Applications; SIAM
      Review; Math. Program.
Vice-chair: SIAM Activity Group on Linear Algebra.
Referee: SIAM J. Optimization, SIAM J. Matrix An. App., Linear Alg. App., Computational Geometry Theory and App.; Math. Progr.; International Meshing Roundtable; J. Complexity.; NSF Panelist, Numeric, Symbolic and Geometric Computing Program.

Lectures
Sparse Matrices and Automatic Differentiation in Semidefinite Programming, International Conference on
      Advances in Convex Analysis and Global Optimization, Samos, Greece (June 6, 2000).
-. 17th International Symposium on Mathematical Programming, Atlanta, GA (August 7, 2000).

Publications
"Quality Mesh Generation in Higher Dimensions." SIAM J. Comput. 29:1334-1370 (2000). With S. A. Mitchell.
"Accurate Solution of Weighted Least Squares by Iterative Methods." SIAM J. Matrix An. App. 22:1153-1174
      (2001). With E. Y. Bobrovnikova.

Werner Vogels
Research Associate
vogels@cs.cornell.edu
http://www.cs.cornell.edu/vogels
Ingenieur Hogere Informatica
Haagse Hogeschool, 1989

My research explores the impact of scale on reliable distributed systems. The main focus is on the development of new network protocols and middleware, as well as on novel strategies to structure applications and support systems.
In the context of the Spinglass project, I am collaborating with Ken Birman and Robbert van Renesse on the development of a new generation of high-scalable reliable network protocols based on the principles of epidemic information dissemination. Although the research has already resulted in protocols for reliable multicast and group membership and failure detection, there are still many open questions such as the application of epidemic techniques to congestion control for many-to-many multicast protocols.
In the Galaxy project, the focus is on the distributed systems needs of enterprise cluster computing systems. In particular, I investigate the scalability problems that arise in these systems and try to provide solutions in a form that is directly applicable to current practical problems. Currently I am looking at the scalable cluster problems from three angles: foremost I try to address cluster management scaling problems by developing a framework for managing complete cluster farms with many different styles of cluster computing present in the farm. Secondly, I am investigating some of the issues that arise when providing support for the development of complex cluster aware applications. Most recently, I have started to address some of the more complex cluster system structuring issues such as meta-clusters and geographical distributed clusters.
I also remain active in the field of high-performance cluster communication. In the past, I collaborated with Thorsten von Eicken on building high-performance user-level network interfaces, which eventually resulted in the VIA industry standard for user-level network interfaces. More recently, I have started investigating the usage patterns of VIA based devices in large-scale clusters, especially with respect to hot-spot control and high-performance flow control.
PROFESSIONAL ACTIVITIES
Steering Committee: Usenix Windows Systems Symposium.
Program Committee: Usenix 4th Windows Systems Symposium, IEEE International Conference on Cluster Computing-Cluster 2000; 2001 IEEE Symposium on Applications and the Internet; International SRDS Workshop on Dependable System Middleware and Group Communication; Usenix 6th Conference on Object Oriented Tools and Systems (Program Committee); IEEE 2nd Workshop on Internet Applications-WIAPP'01; IFIP/ACM International Conference on Distributed Systems Platforms - Middleware 2001.
Program Chair: Distributed Systems Track of 2001IEEE Symposium on Applications and the Internet.
General Chair: IEEE 2nd Workshop on Internet Applications-WIAPP'01.

LECTURES
Cluster Computing Made Easy: New Tools for Scalable Servers and Services." Invited Lecture, Annual meeting of Advanced Cluster Computing Consortium (June 2, 2000).

PUBLICATIONS
"Spinglass, Scalable and Secure Communication Tools for Mission-critical Computing." Proceeedings of the 2001 DARPA Information Survivability Conference and Exhibition - II (June, 2001). With K. Birman, and R. van Renesse.
"Using Epidemic Techniques for Building Ultra-scalable Reliable Communication Systems." Proceedings of the Large Scale Networking Workshop: Research and Practice, Vienna, VA (March, 2001). With K. Birman and R. van Renesse.
"An Overview of the Galaxy Management Framework for Scalable Enterprise Cluster Computing." Proceedings of the IEEE International Conference on Cluster Computing: Cluster-2000, Chemnitz, Germany (December, 2000). With D. Dumitriu.
"Tree-saturation in the AC3 Velocity Cluster Interconnect." Proceedings of the 8th conference on Hot Interconnects, Stanford, CA (August, 2000). With D. Follett, J. Hsieh, D. Lifka, and D. Stern.

Golan Yona
Assistant Professor
golan@cs.cornell.edu
http://www.cs.cornell.edu/golan
Ph.D. Hebrew University, Jerusalem, 1999

My research focuses on computational molecular biology, with an emphasis on developing tools and methodologies for large-scale analysis of protein sequences and structures.
     The goal of my research is to explore high-order organization within the space of all proteins and obtain a global view (a "road map") of the protein space. We hope that the global view will yield valuable insights about the nature and function of new genes and will lead to the discovery of high-level properties and principles in the protein space.
     This interdisciplinary research is rooted in two different disciplines: computer science and molecular biology. Being on the borderline between the two disciplines, this study is related to fields of intensive research in both. It incorporates study and development of methods for metric embedding, unsupervised learning techniques, efficient graph algorithms and parallel applications, and efficient database management. On the computational biology side, it is involved with development of new algorithms and approaches for protein comparison, statistical models of protein families and study of the mapping from sequences to structures. A great emphasis is on developing novel machine-learning based techniques, both in the context of the study of the protein space and as general-purpose tools.
     My study so far has resulted in two large databases that are being used by biologists to study new genes: ProtoMap: this database and its interactive web sites (http://protomap.cornell.edu and http://protomap.stanford.edu) are based on a graph-theoretic approach for the large-scale organization of protein sequences. The site offers an automatic hierarchical classification of all known protein sequences. It is a useful resource for the analysis of known as well as new protein sequences and for the study of relationships between protein families.
     BioSpace: This database and its web sites (http://biospace.cornell.edu and http://biospace.stanford.edu) are based on a new unified framework for sequence and structure analysis. The result of this analysis is a preliminary global map of the protein space. The sites also provide all-atom three-dimensional models for over 160,000 proteins.

Awards
Burroughs-Welcome Fellowship from the Program in Mathematics and Molecular Biology (PMMB) (1998-
      2000).
Noted for excellency in teaching, Faculty of Mathematics and Natural Sciences, Hebrew University, Jerusalem,
      Israel (1998 ).
Faculty of Mathematics and Natural Sciences Research award, Hebrew University, Jerusalem, Israel (1998).
Jewish National KKL Foundation Research award (1996).
Intel-Dean Prize (1996).

Lectures
Towards a Complete Map of the Protein Space based on a Unified Sequence and Structure Analysis of all
      Known Proteins. NIH Symposium: From Genes to Proteins to Biological Function, Ithaca, NY (October,
      2000).
Towards a Complete Map of the Protein Space based on a Unified Sequence and Structure Analysis of all
      Known Proteins. The Eighth International Conference on Intelligent Systems for Molecular Biology (ISMB),
      San Diego, CA (August, 2000).
A Unified Sequence-structure Classification of Protein Sequences: Combining Sequence and Structure in a
      Map of the Protein Space. Fourth Annual International Conference on Computational Molecular Biology
      (RECOMB), Tokyo, Japan (April, 2000).
A Unified Sequence-structure Classification of Protein Sequences: Combining Sequence and Structure in a
      Map of the Protein Space. Quantitative Challenges in the Post-genomic Sequence Era, San Diego, CA
      (January, 2000).
Methodologies for Target Selection in Structural Genomics. Innovative Computational Applications, San
      Francisco, CA (October, 1999).
Modeling Protein Families using Probabilistic Suffix Trees. Third Annual International Conference on
      Computational Molecular Biology (RECOMB), Lyon, France (April, 1999).
A Map of the Protein Space-An Automatic Hierarchical Classification of all Known Proteins. The Sixth
      International Conference on Intelligent Systems for Molecular Biology (ISMB), Montreal, Canada
      (June,1998).

Publications
"Within the Twilight Zone: A Sensitive Profile-profile Comparison Tool based on Information Theory." Journal of
      Molecular Biology (2001). With M. Levitt.
"Variations on Probabilistic Suffix Trees: Statistical Modeling and Prediction of Protein Families.
      Bioinformatics 17:23-43 (2001). With G. Bejerano.
"Towards a Complete Map of the Protein Space based on a Unified Sequence and Structure Analysis of All
      Known Proteins. In the Proceedings of ISMB 2000, 395-406, AAAI Press. With M. Levitt.
"A New Nonparametric Pairwise Clustering Algorithm based on Iterative Estimation of Distance Profiles.
      Machine Learning (2001). With S. Dubnov, R. El-Yaniv, Y. Gdalyahu, E. Schneidman, and N. Tishby.
"Methodologies for Target Selection in Structural Genomics. Progress in Biophysical and Molecular Biology
      73:297-320 (2000). With M. Linial.
"A Unified Sequence-structure Classification of Protein Sequences: Combining Sequence and Structure in a
      Map of the Protein Space. Proceedings of RECOMB 2000, 308-317, ACM press. With M. Levitt.
"Comparison of Protein Sequences and Practical Database Searching. In BioInformatics: Sequence, Structure,
      and Databanks, edited by D. Higgins and W. Taylor. Oxford University Press (2000). With S. Brenner.

Landmark Publications
"ProtoMap: Automatic Classification of Protein Sequences, a Hierarchy of Protein Families, and Local Maps of
      the Protein Space. Proteins: Structure, Function and Genetics 37:360-378 (1999). With N. Linial, and M.
      Linial.
"Global Self-organization of All Known Protein Sequences Reveals Inherent Biological Signatures. Journal of
      Molecular Biology 268:539-556 (1997). With M. Linial, N, Linial, and N. Tishby.

Ramin Zabih
Associate professor
rdz@cornell.edu
http://www.cs.cornell.edu/rdz/
Ph.D. Stanford University, 1994

My research interests lie in computer vision and in medical imaging. I have worked on a variety of problems in early vision, including motion and stereo; many of these problems can be solved very accurately using algorithms based on graph cuts. In the last year I have been doing research on medical imaging in the Radiology Department at Cornell Medical School, where I have focused on improving the quality of MRIs. I have also investigated a number of applications of computer vision, including new methods for content-based access to databases of images, and have also developed some simple computer vision techniques to automate program debugging at Microsoft.

Lectures
Fast Approximate Energy Minimization via Graph Cuts. University of Washington CS Colloquium (January,
      2001).

Publications
"Visual Correspondence with Occlusions via Graph Cuts." Proc. International Conference on Computer Vision
      (ICCV) (July, 2001). With Vladimir Kolmogorov.
"An Efficient Real-time Navigator Algorithm: Motion Organized Simultaneous Acquisition with Interactive
      Control (MOSAIC)." Proc. IEEE Conference on Medical Imaging and Augmented Reality, Hong Kong
      (June, 2001). With Vladimir Kolmogorov, Yi Wang, Richard Watts, and Martin Prince.
"Exact Voxel Occupancy with Graph Cuts." Proc. IEEE Conference on Computer Vision and Pattern
      Recognition (CVPR) (2000). With Dan Snow, and Paul Viola.

Landmark Publications
"Color-spatial Indexing and Applications." International Journal of Computer Vision 35(3) (1999). With Jing
      Huang, S. Ravi Kumar, Mandar Mitra, and Wei-Jing Zhu.
"Fast Approximate Energy Minimization via Graph Cuts." Proc. International Conference on Computer Vision
      (ICCV) (1999). With Yuri Boykov, and Olga Veksler.
"Markov Random Fields with Efficient Approximations." Proc. IEEE Conference on Computer Vision and
      Pattern Recognition (CVPR) (1998). With Yuri Boykov, and Olga Veksler.
"A Variable Window Approach to Early Vision." IEEE Transactions on Pattern Analysis and Machine
      Intelligence 20(12) (1998). With Yuri Boykov, and Olga Veksler.