Artificial Intelligence

Home » Research

"I could feel, I could smell, a new kind of intelligence across the table." --Garry Kasparov

A Banner Year for AI at Cornell!

Understanding intelligence and creating intelligent artifacts, the twin goals of artificial intelligence (AI), represent two of the final frontiers of modern science. Several of the early pioneers of computer science, such as Turing, Von Neumann, and Shannon, were captivated by the idea of creating a form of machine intelligence. The questions and issues considered back then are still relevant today, perhaps even more so.

Research in AI at Cornell covers a wide range of topics, including decision theory, knowledge representation, machine learning and datamining , natural language processing , planning, reasoning under uncertainty, robotics, search, and vision . Our research program embraces both theoretical and experimental aspects - a particular strength of the department is in compute-intensive approaches to AI problems. Given the complexity of many of the basic questions in AI, our research often transcends traditional scientific boundaries. We are actively pursuing connections to other disciplines, such as biology, economics, linguistics, operations research, physics, and psychology. The Department is also one of the main participants in the university-wide Cognitive Science program and in the Intelligent Information Systems Institute , a collaboration with the Air Force Research Laboratory.

Selected faculty profiles

Claire Cardie works on natural language understanding and machine learning, developing corpus-based techniques that allow a system to bootstrap its own knowledge bases directly from text. Cardie's decision tree-based approach to feature selection is in use in IBM's commercial data-mining products. She was a co-developer of the highest-ranked information extraction system at the DARPA-sponsored Third Message Understanding Conference (MUC-3). Cardie won an NSF CAREER Award in 1996.

Carla Gomes works on solving hard combinatorial problems by combining techniques from artificial intelligence and operations research. She focuses on studying the role of randomization in computation, characterization of the distribution profiles (especially "heavy-tailed distributions") of randomized algorithms, and consequences for algorithm design. Gomes won the Best Consultant award from the Information Directorate of the Air Force Research Laboratory, co-won the Distinguished Paper Award at the 2004 Conference on the Principles and Practice of Constraint Programming (CP-2004), and is the director of the Intelligent Information Systems Institute .

Joe Halpern works on reasoning about knowledge and uncertainty, with applications to distributed computing and game theory. He has also done work in, and is still interested in, modal logic, security, fault tolerance, and resource-bounded reasoning. Halpern is a Fellow of the ACM, a Fellow of the AAAS, and AAAI and has received many awards, including two best paper awards at IJCAI (Intern. Joint Conference on AI), a best paper award at KR 2006 (Intern. Conference on Principles of Knowledge Representation and Reasoning), a Guggenheim Fellowship, a Fulbright Fellowship, and the 1997 Gödel Prize.

Dan Huttenlocher works on visual matching and recognition in computer vision. His research interests also include new types of electronic documents for communication, collaboration, and education, and e-commerce.

Thorsten Joachims ' research interests center on a synthesis of theory and system building in machine learning, with a focus on Support Vector Machines and machine learning with natural language text (e.g. text classification, information retrieval). He is the author of the well-known SVM-light package for Support Vector Machines. He won an NSF CAREER award in 2003, Best Paper Award at ICML 2005, and Best Research Paper Award at KDD 2006.

Lillian Lee works in the area of natural language processing, focusing on "knowledge-lean" methods for automatically learning linguistic knowledge from essentially raw text. She received an Alfred P. Sloan Research Fellowship in 2002 and in 2004 was the co-recipient of the first Best Paper award from HLT-NAACL. A joint project on corpus-based paraphrase induction was described in this New York Times article .

Hod Lipson works on Robotics and embodied AI, with a focus on active learning methods where machines actively probe their surroundings to learn what they need. He also works in symbolic machine learning techniques for complex dynamical systems, from social systems to gene regulation networks. He directs the Computational Synthesis Lab that looks at new ways to design and adapt systems using mostly biologically-inspired approaches. Lipson received an NSF CAREER Award in 2006 and a DARPA Young Faculty Award in 2007.

Bart Selman 's current research interests include efficient reasoning procedures, stochastic search methods, theory approximation, knowledge compilation, planning, software agents, and connections between computer science, operations research and physics. Selman has received a number of best paper awards, an NSF CAREER award, and an Alfred P. Sloan Research Fellowship, and is a Fellow of AAAI and the American Association for the Advancement of Science. You can read about his phase transition work in a New York Times article .

Ramin Zabih works on computer vision and medical imaging. He has studied a variety of problems in early vision, and has investigated solutions employing graph cuts. He holds a joint appointment with the Radiology department at Weill Cornell Medical School. He won two Best Paper awards at ECCV 2002.

Affiliated faculty

Claire Cardie
Rich Caruana
Shimon Edelman
Carla Gomes
Joe Halpern
Dan Huttenlocher
Thorsten Joachims
Lillian Lee
Hod Lipson
Mats Rooth
Bart Selman
Phoebe Sengers
Ramin Zabih