|
Department
of |
|
Susan L. EpsteinThe CUNY Graduate School, Department
of Computer Science and Hunter College, Department
of Computer Science |
|||||
|
|
|
|
|||||
|
Home Publications Collaborators Courses Contact
|
|
ACE (CSP)The Adaptive Constraint EngineACE 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. Current
Work Crucial
subproblems and structure Learning
inference methods Reinforcement
learning for ranked preferences. Subgoals for efficient, effective
search.
Key
references Petrovic, S. and S. L. Epstein. 2006. Full Restart Speeds Learning. In Proceedings
of FLAIRS-2006. 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. 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. #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. |
