Research Highlights

       

Dimensionality Reduction with Complex Constraints for Scientific Discovery: Application to High-throughput Materials Discovery

  • In this research, we solve the phase-map identification problem to determine the crystal structure of materials based on high-energy synchrotron-based X-ray diffraction (XRD) data. Our AI solution tightly integrates machine learning, automated reasoning, as well as crowdsourcing and human computation.
  • Since our AI platform has been deployed at the Department of Energy's Joint Center for Artificial Photosynthesis (JCAP), thousands of X-ray diffraction patterns have been processed and the results yield the discovery of new materials for energy applications.
  • Our scientific discovery is featured as editor's choice and the cover story in American Chemical Society Combinatorial Science, and received the IAAI Innovative Application Award.
  • Papers: [C16] [J3] [C14] [C9] [C6].
                       
       

Avicaching: a Two Stage Game for Bias Reduction in Citizen Science

  • In this research, we introduce Avicaching as a game theoretic solution to address the data bias problem in citizen science. Avicaching is a novel two-stage game, in which the organizer, a citizen-science program, incentivizes the agents, the citizen scientists, to visit under-sampled locations.
  • We provide OPTIKA, a novel way of encoding this two-stage game as a single optimization problem, cleverly embedding (an approximation of) the agents' problems into the organizer's problem. When implemented in the eBird citizen science program, our Avicaching game shifted 19% birding effort from traditional hotspots to undersampled locations in 3 counties in upstate New York.
  • Our story is featured in NSF news. [video][Avicaching Website] Papers: [C10] [C12] [C5].
                       
            

Solving Marginal MAP problems with NP Oracles and XOR Constraints

  • We solve the Marginal Maximum A Posteriori (Marginal MAP) problem, arising naturally from many applications at the intersection of decision-making and machine learning. Marginal MAP problems unify the two main classes of inference, namely maximization (optimization) and marginal inference (counting).
  • Our solution embeds the intractable counting subproblem as queries to NP oracles, subject to additional XOR constraints, leading to a constant factor approximation algorithm to solve the Marginal MAP problem. Our approach has been applied in probabilistic reasoning (Upper-left in the picture on the left), machine learning (Upper-right), information cascade and network design (Bottom).
  • [spotlight video] Papers: [C13] [C15].
                       
            

Compact Knowledge Representation in the Fourier Domain

  • In this research, we present a compact representation of high dimensional knowledge based on discrete Fourier representations, complementing the classical approach based on conditional independence. We show a large class of probabilistic graphical models have a compact Fourier representation. We demonstrate the significance of this representation by applying it to the variable elimination algorithm. We show that a simple algorithm with a new representation leads to competitive scores on UAI inference challenge instances.
  • Paper: [C11].

Publications

Conference Papers & Journal Articles

[C16]   Yexiang Xue, Junwen Bai, Ronan Le Bras, Brendan Rappazzo, Richard Bernstein, Johan Bjorck, Liane Longpre, Santosh K. Suram, Robert B. van Dover, John Gregoire, and Carla Gomes.
Phase-Mapper: An AI Platform to Accelerate High Throughput Materials Discovery.
In Proc. 29th Annual Conference on Innovative Applications of Artificial Intelligence (IAAI), 2017. [pdf][video 1][video 2][video 3]
IAAI Innovative Application Award

[C15]   Yexiang Xue, Xiaojian Wu, Dana Morin, Bistra Dilkina, Angela Fuller, J. Andrew Royle, and Carla Gomes.
Dynamic Optimization of Landscape Connectivity Embedding Spatial-Capture-Recapture Information.
In Proc. 31th AAAI Conference on Artificial Intelligence (AAAI), 2017. [pdf] [supplementary materials]

[J3]   Santosh K. Suram, Yexiang Xue, Junwen Bai, Ronan LeBras, Brendan H Rappazzo, Richard Bernstein, Johan Bjorck, Lan Zhou, R. Bruce van Dover, Carla P. Gomes, and John M. Gregoire.
Automated Phase Mapping with AgileFD and its Application to Light Absorber Discovery in the V-Mn-Nb Oxide System.
In American Chemical Society Combinatorial Science, Dec, 2016. [DOI][pdf][video 1][video 2][video 3]
Editor's choice and the cover story!

[C14]   Junwen Bai, Johan Bjorck, Yexiang Xue, Santosh K. Suram, John Gregoire, and Carla Gomes.
Relaxation Methods for Constrained Matrix Factorization Problems: Solving the Phase Mapping Problem in Materials Discovery.
To appear in the Fourteenth International Conference on Integration of Artificial Intelligence and Operations Research Techniques in Constraint Programming (CPAIOR), 2017.

[C13]   Yexiang Xue, Zhiyuan Li, Stefano Ermon, Carla P. Gomes, Bart Selman.
Solving Marginal MAP Problems with NP Oracles and Parity Constraints
In the Proceedings of the 29th Annual Conference on Neural Information Processing Systems (NIPS), 2016. [pdf] [spotlight video]

[C12]   Yexiang Xue, Ian Davies, Daniel Fink, Christopher Wood, Carla P. Gomes.
Behavior Identification in Two-stage Games for Incentivizing Citizen Science Exploration
In the Proceedings of the 22nd International Principles and Practice of Constraint Programming (CP), 2016. [pdf][video]
** Click [here] to participate in the fun Avicaching Game!

[C11]   Yexiang Xue, Stefano Ermon, Ronan Le Bras, Carla P. Gomes and Bart Selman.
Variable Elimination in the Fourier Domain
In the Proceedings of the 33rd International Conference on Machine Learning (ICML), 2016. [pdf][supplementary materials][video in Simons Institute]

[C10]   Yexiang Xue, Ian Davies, Daniel Fink, Christopher Wood, Carla P. Gomes.
Avicaching: A Two Stage Game for Bias Reduction in Citizen Science
In the Proceedings of the 15th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2016. [pdf][supplementary materials][video]
** Click [here] to participate in the fun Avicaching Game!

[C9]   Yexiang Xue, Stefano Ermon, Carla P. Gomes, Bart Selman.
Uncovering Hidden Structure through Parallel Problem Decomposition for the Set Basis Problem: Application to Materials Discovery.
In the Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence (IJCAI), 2015. [pdf] [supplementary materials][video]

[C8]   Stefano Ermon, Yexiang Xue, Russell Toth, Bistra Dilkina, Richard Bernstein, Theodoros Damoulas, Patrick Clark, Steve DeGloria, Andrew Mude, Christopher Barrett, and Carla Gomes.
Learning Large Scale Dynamic Discrete Choice Models of Spatio-Temporal Preferences with Application to Migratory Pastoralism in East Africa.
In Proc. 29th AAAI Conference on Artificial Intelligence (AAAI), 2015. [pdf]

[C7]   Yilun Wang, Yu Zheng, and Yexiang Xue.
Travel Time Estimation of a Path using Sparse Trajectories.
In the Proceeding of the 20th SIGKDD conference on Knowledge Discovery and Data Mining (KDD), 2014. [pdf]

[C6]   Ronan Le Bras, Yexiang Xue, Richard Bernstein, Carla P. Gomes, Bart Selman.
A Human Computation Framework for Boosting Combinatorial Solvers.
In Second AAAI Conference on Human Computation and CrowdSourcing (HComp), 2014. [pdf]

[C5]   Yexiang Xue, Bistra Dilkina, Theodoros Damoulas, Daniel Fink, Carla P. Gomes and Steve Kelling.
Improving Your Chances: Boosting Citizen Science Discovery.
In First AAAI Conference on Human Computation and CrowdSourcing (HComp), 2013. [pdf] [hot spot list] [species list].

[J2]   Stefano Ermon, Yexiang Xue, Carla Gomes, and Bart Selman.
Learning Policies For Battery Usage Optimization in Electric Vehicles.
In Machine Learning (ML), 2013. [online version]

[C4]   Ronan Le Bras, Bistra Dilkina, Yexiang Xue, Carla P. Gomes, Kevin S. McKelvey, Claire Montgomery and Michael K. Schwartz.
Robust Network Design for Multispecies Conservation.
In Proceedings of the Twenty-Seventh AAAI Conference on Artificial Intelligence (AAAI), 2013. [pdf]

[C3]   Bistra Dilkina, Katherine Lai, Ronan Le Bras, Yexiang Xue, Carla P. Gomes, Ashish Sabharwal, Jordan Suter, Kevin S. McKelvey, Michael K. Schwartz and Claire Montgomery.
Large Landscape Conservation - Synthetic and Real-World Datasets.
In Proceedings of the Twenty-Seventh AAAI Conference on Artificial Intelligence (AAAI), 2013. [pdf]

[C2]   Yexiang Xue, Arthur Choi, and Adnan Darwiche.
Basing Decisions on Sentences in Decision Diagrams.
In Proceedings of the Twenty-Sixth AAAI Conference on Artificial Intelligence (AAAI), 2012. [pdf]

[C1]   Stefano Ermon, Yexiang Xue, Carla Gomes, and Bart Selman.
Learning Policies For Battery Usage Optimization in Electric Vehicles.
In In Proceedings of European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD),2012 [pdf]

[J1]   Arthur Choi, Yexiang Xue, and Adnan Darwiche.
Same-Decision Probability: A Confidence Measure for Threshold-Based Decisions.
In the International Journal of Approximate Reasoning (IJAR), Vol. 53, No. 9, 2012. [pdf]

Working Manuscripts

[W1]   Di Chen, Yexiang Xue, Shuo Chen, Daniel Fink, and Carla P. Gomes.
Deep Multi-species Embedding.
[Arxiv].

Education

  • September, 2011 -- Now, PhD student, Inst. of Computational Sustainability & Computer Science Dept., Cornell University.

  • September, 2007 -- July, 2011, Undergraduate, EECS, Peking University.

  • Feburary, 2011 -- May, 2011, July, 2010 -- September, 2010, Research Assistant, Automated Reasoning Lab, UCLA, Mentor: Prof. Adnan Darwiche.

Professional Activities

PC Member: AAAI-17, UAI-17.
Reviewer for AAAI, IJCAI, UAI, NIPS, KDD, CP, CPAIOR.

Undergraduate students mentored:

  • Zhiyuan Li   (Considering multiple PhD offers from top-5 US institutions.)
  • Junwen Bai   (Committed to be a PhD student in Cornell)
  • Di Chen   (Committed to be a PhD student in Cornell)

Media Coverage

Computational Sustainability

Computers play a crucial role in preserving the Earth. NSF News - 4/20/2016

Big data experts to share green ideas at World Economic Forum. Cornell Chronicle - 6/24/2016

Combinatorial Materials Discovery

Materials to do anything under the sun Cornell Engineering Magazine - 10/4/2016

eBird Citizen Science Program & Avicaching Incentive Game

Understanding birders to better understand birds North American Ornithological Conference - 08/16/2016

Computational Sustainability for Everyone: Untapping the Potential of Games, As Told by Pokémon GO Computational Sustainability Blog - 07/18/2016

3 ways artificial intelligence will save the day GreenBiz - 6/27/2016

Three ways artificial intelligence is helping to save the world Ensia - 4/26/2016

Incentivizing citizen science discovery for a sustainable world Phys.org - 2/13/2016

Wildlife Corridor Preservation

Computing cost-effective wildlife corridors Monabay News, 11/11/2016

When animals share, conservation is affordable Cornell Chronicle - 10/27/2016

Optimization technique identifies cost-effective biodiversity corridors ScienceDaily - 9/27/2016

Ecological corridor to preserve Ecuadorian Andes bears Cornell Chronicle - 3/9/2015

Forging a New Path: Working to Build the Perfect Wildlife Corridor Pacific Standard Nature & TEch - 9/25/2014

Forging a New Path On Earth - 9/15/2014

Big Data for African Herders

Economist, partners clinch USAID award for drought insurance Cornell Chronicle - 10/12/2016

App tracks Kenya's best places to graze Futurity Science and Technology - 2/20/2015

Space-age technology points African herders in right direction Cornell Chronicle - 2/15/2015

Work

  • May, 2013 -- August, 2013, Research Intern, Microsoft Research Asia, Beijing, Mentor: Yu Zheng.

Institute of Computational Sustainability, Cornell University. Last Modified, March, 2017.