CORE

                 Computational Principles  for Optimization of Resources and Execution Time

Core Home
UW CS
Cornell CS
MS Research

Research team

Henry Kautz
Eric Horvitz
Bart Selman
Carla Gomes
Yongshao Ruan


 

 

Overall Goal and Techniques

The CORE project is developing new approaches to speeding up reasoning and search by synthesizing ideas from the study of phase transition phenomena in problem distributions, decision-theoretic control of reasoning, Bayesian inference, and machine learning.  Current research centers on the construction of probabilistic models that can predict the run time of problem solving algorithms, and the development of solver control strategies that take advantage of such predictive models to improve performance.  Work from the CORE project is showing how large, hard search and reasoning problems can be solved in practice by leveraging the inherent uncertainty and variability in runtime of combinatorial algorithms.
 
 

References

1 Eric Horvitz, Yongshao Ruan, Carla P. Gomes, Henry Kautz, Bart Selman, David Maxwell Chickering.
A Bayesian Approach to Tackling Hard Computational Problems
Proceedings the 17th Conference on Uncertainty in Artificial Intelligence (UAI-2001). 
 
2 Henry Kautz, Yongshao Ruan, Dimitiri Achlioptas, Carla Gomes, Bart Selman, and Mark Stickel. 
Balance and Filtering in Structured Satisfiable Problems. 

Proceedings of the 17th International Joint Conference on Artificial Intelligence (IJCAI-2001). 

3 Dimitris Achlioptas, Carla Gomes, Henry Kautz, and Bart Selman. 
Generating Satisfiable Problem Instances
Proceedings of  the 17th National Conference on Artificial Intelligence (AAAI-2000). 
 

Core Home
UW CS
Cornell CS
MS Research

Presentations

A Bayesian Approach to Tackling Hard Computational Problems
Eric Horvitz, presented at the the 17th Conference on Uncertainty in Artificial Intelligence (UAI-2001). 

Learning to Search. 
Henry Kautz, presented at the Workshop on Stochastic Search, IJCAI-2001, Seattle, WA. 
 

Links

Internal CORE Project page (password protected)

TASK Project at Cornell

Project Support

This material is based upon work supported by the National Science Foundation  under Grant No. 0120307.  Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

We also gratefully acknowledge support from the Microsoft Corporation, and the Intelligent Information Systems Institute.

IISI Logo
NSF Home Page

National
Science
Foundation

Microsoft Home