ACE (CSP)
The Adaptive Constraint Engine
ACE is a FORR-based program that
learns to solve constraint satisfaction problems. Its premise is that
agreement among varying heuristic viewpoints is a valid decision-making
principle. ACE minimizes search, focusing instead upon reasonable rationales
and multiple learning methods. ACE learns during problem solving, and
demonstrates substantial, learned expertise after relatively little
training.
Available
problem classes
Problem
Solving and Machine Learning Laboratory
Key
references
Li, X. and S. L. Epstein (2010). Learning cluster-based structure to solve constraint satisfaction problems, Annals of Mathematics and Artificial Intelligence..
Epstein, S. L. and X. Yun 2010. From Unsolvable to Solvable: An Exploration
of Simple Changes. In Proceedings of WARA-10.
Li, X. and S. L. Epstein 2010. Visualization for Structured Constraint
Satisfaction Problems. In Proceedings of AAAI Workshop on Visual
Representations and Reasoning.
Epstein, S. L., & Petrovic, S. (In press). Learning Expertise with Bounded Rationality and Self-awareness. In Metareasoning: Thinking about thinking: MIT Press.
Epstein,
S. L. (2009). Integrating a Portfolio of Representations to Solve Hard
Problems. In Proceedings of the AAAI Fall Symposium on Multi-representational
Architectures for Human-level Intelligence.
Epstein,
S. L. and X. Li (2009). Cluster-based Modeling for Constraint Satisfaction
Problems. In Proceedings of the IJCAI Workshop on Learning Structural
Knowledge from Observations.
Epstein,
S. L., & Li, X. (2009). Cluster Graphs as Abstractions for Constraint
Satisfaction Problems. In Proceedings of SARA-09.
Epstein,
S. L. and X. Li (2009). Search on Constraint Satisfaction Problems with
Sparse Secondary Structure. In Proceedings of International Symposium
on Combinatorial Search (SoCS-09).
Petrovic,
S. and S. L. Epstein (2008). Tailoring a Mixture of Search Heuristics.
Constraint Programming Letters 4: 15-38.
Epstein,
S. L. (2008). Building a Constraint Solver that Learns. In Proceedings of the AAAI Fall Symposium on BIologically Inspired Computer Architecture, Arlington VA. AAAI.
Epstein, S. L. (2008). Optimistic Problem Solving. In Proceedings of the AAAI
Fall Symposium on Naturally Inspired Artificial Intelligence, Arlington VA.
Epstein,
S. L. and S. Petrovic (2008). Learning Expertise with Bounded Rationality
and Self-awareness. In Proceedings of AAAI Workshop on Metareasoning,
Chicago, AAAI.
Zhang,
Z. and S. L. Epstein (2008). Learned Value-Ordering Heuristics for Constraint
Satisfaction. In Proceedings of STAIR-08 Workshop at AAAI-2008.
Petrovic,
S. and S. L. Epstein (2007). Random Subsets Support Learning a Mixture
of Heuristics. International Journal on Artificial Intelligence Tools
20(10): 1-17.
(An earlier,
less detailed version appeared as Petrovic, S. and S. L. Epstein (2007).
Learning to Solve Constraint Problems. ICAPS-07 Workshop on Planning
and Learning, Providence RI.)
Petrovic,
S. and S. L. Epstein (2007). Random Subsets Support Learning a Mixture
of Heuristics. In Proceedings of FLAIRS (2007), Key West, AAAI.
Petrovic,
S., S. L. Epstein and R. J. Wallace (2007). Learning a Mixture of Search
Heuristics. In Proceedings of CP-07 Workshop on Autonomous Search, Providence,
RI.
Petrovic,
S. and S. L. Epstein (2007). Preferences Improve Learning to Solve Constraint
Problems. AAAI-07 Workshop on Preference for Artificial Intelligence.
Zhang,
Z. and S.L. Epstein, (2007). Constraint Solving by Composition. Proceedings
of CP-07, Providence, RI.
Epstein, S. L. and R. J. Wallace. 2006. Finding Crucial Subproblems to
Focus Global Search. In Proceedings of ICTAI-2006, Washington, D.C., IEEE.
Petrovic, S. and S. L. Epstein. 2006. Full Restart Speeds Learning. In Proceedings
of FLAIRS-2006.
Epstein, S. L., E. C. Freuder and M. Wallace 2005. Learning to Support
Constraint Programmers. Computational Intelligence 21(4): 337-371.
Epstein, S. L., E. C. Freuder, R. M. Wallace and X. Li. 2005. Learning
Propagation Policies. In Proceedings of the Second International Workshop
on Constraint Propagation and Implementation,
Sitges, Spain, pp.1-15.
Epstein, S. L. and T. Ligorio. 2004. Fast and Frugal Reasoning Enhances a
Solver for Really Hard Problems. In Proceedings of Cognitive Science 2004. Chicago: Lawrence Earlbaum, pp.351-356
Epstein, S. L.,
E.C. Freuder, R. Wallace, A. Morozov and B. Samuels. 2002. The Adaptive
Constraint Engine. In Principles and Practice of Constraint Programming --
CP2002, 2470. Berlin: Springer Verlag.
Epstein, S. L.
and G. Freuder. 2001. Collaborative Learning for Constraint Solving. In Principles
and Practice of Constraint Programming -- CP2001, 2239. Berlin: Springer Verlag.
Research
on ACE is done in collaboration with Gene Freuder, Rick Wallace, and
the Cork Constraint Computation Centre.
This
material is based upon work supported by the National Science Foundation
under Grant Nos. ISS-0811437, IIS-0328743, 9423085, #IRI-9703475, 9222720,
and #9001936, by the New York State Technological Development Graduate
Research and Technology Initiative, and by the PSC-CUNY Research Foundation.
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, New York State, or PSC-CUNY.
[top
of the page]
|
