Learning from opportunities: storing and re-using execution-time optimizations
作者： Kristian Hammond , Tim Converse , Mitchell Marks
DOI:
关键词: Execution time 、 Service (systems architecture) 、 Software engineering 、 Artificial intelligence 、 Work (electrical) 、 Set (abstract data type) 、 Domain (software engineering) 、 Computer science 、 Plan (drawing)
摘要: In earlier work (Hammond 1986), we proposed a mechanism for learning from execution-time plan failure. this paper, suggest corollary notion of planning opportunities. We argue that both are special cases expectation failure (Schank 1982). The result type is set plans frequently occurring conjuncts goals, indexed by the features in world predict their usefulness. discuss notion, using examples University Chicago planner TRUCKER, an implementation case-based domain UPS-like pickup and delivery service.
参考文章(12)
Austin Tate, INTERPLAN: a Plan Generation System Which Can Deal With Interactions Between Goals University of Edinburgh Department of Machine Intelligence and Perception. ,(1974)
Kristian J. Hammond, Learning to anticipate and avoid planning problems through the explanation of failures national conference on artificial intelligence. pp. 556- 560 ,(1986)
Philip E. Agre, David Chapman, Pengi: an implementation of a theory of activity national conference on artificial intelligence. pp. 268- 272 ,(1987)
Thomas Dean, David Miller, R J Firby, The Forbin Paper. ,(1987)
Schank Roger C, Dynamic Memory: A Theory of Reminding and Learning in Computers and People ,(1983)
Jaime G. Carbonell, Learning by Analogy: Formulating and Generalizing Plans from Past Experience Machine Learning. pp. 137- 161 ,(1983) , 10.1007/978-3-662-12405-5_5
Michael Lebowitz, Generalization and memory in an integrated understanding system Yale University. ,(1980)
Lawrence Albert Birnbaum, Integrated processing in planning and understanding Unknown Publisher. ,(1986)
Symposium on the Fetus and Birth (1976 : Ciba Foundation), A Structure for Plans and Behavior ,(1977)
Gerald J. Sussman, A Computational Model of Skill Acquisition Massachusetts Institute of Technology. ,(1973)