Learning by Analogical Replay in Prodigy: First Results
[Technical Report, Research Report]
Carnegie Mellon University
Pagination or Media Count:
Robust reasoning requires learning from problem solving episodes. Past experience must be compiled to provide adaptation to new contingencies and intelligent modification of solutions to past problems. This paper presents a comprehensive computational model of analogical reasoning that transitions smoothly between case reply, case adaptation, and general problem solving, exploiting and modifying past experience when available and resorting to general problem-solving methods when required.