FORR
is a satisficing architecture for learning and
problem solving. Its premise is that agreement among
varying heuristic viewpoints is a valid
decision-making principle. A FORR-based program
learns to specialize general domain expertise into
problem-class specific expertise. An implemented FORR
shell supports domain-specific development. FORR
explores the relationship among reactivity,
heuristics, and search. Applications include Hoyle
for game playing and Ariadne for robot path finding.
Each of them has demonstrated substantial, learned
expertise after relatively little training. Susan L. Epstein
Current
Work
The integration of visual perception, limited search, and planning with other high-level
reasoning.
Multiagent solutions to
multiple goal problems with (CD)
.
Application
Domains
Key
references
Epstein, S. L. 1994. For the Right Reasons:
The FORR Architecture for Learning in a Skill Domain.
Cognitive Science, 18 (3): 479-511.
Additional
references
Epstein, S. L. (1995). On Heuristic
Reasoning, Reactivity, and Search. In Proceedings of
the Fourteenth International Joint Conference on
Artificial Intelligence, 454-461. Montreal: Morgan
Kaufmann.
Epstein, S. L., Gelfand, J. and Lesniak, J.
1996. Pattern-Based Learning and Spatially-Oriented
Concept Formation with a Multi-Agent, Decision-Making
Expert. Computational Intelligence, 12 (1): 199-221.
