Investigating Explanation-Based LearningGerald DeJong Explanation-Based Learning (EBL) can generally be viewed as substituting background knowledge for the large training set of exemplars needed by conventional or empirical machine learning systems. The background knowledge is used automatically to construct an explanation of a few training exemplars. The learned concept is generalized directly from this explanation. The first EBL systems of the modern era were Mitchell's LEX2, Silver's LP, and De Jong's KIDNAP natural language system. Two of these systems, Mitchell's and De Jong's, have led to extensive follow-up research in EBL. This book outlines the significant steps in EBL research of the Illinois group under De Jong. This volume describes theoretical research and computer systems that use a broad range of formalisms: schemas, production systems, qualitative reasoning models, non-monotonic logic, situation calculus, and some home-grown ad hoc representations. This has been done consciously to avoid sacrificing the ultimate research significance in favor of the expediency of any particular formalism. The ultimate goal, of course, is to adopt (or devise) the right formalism. |
Contents
Explanation Generalization in EGGS | 20 |
Generalizing Explanation Structures | 60 |
Recoverable Simplifications and the Intractable | 128 |
Copyright | |
4 other sections not shown
Other editions - View all
Common terms and phrases
achieve acquired rules actions ADEPT algorithm analyzing antecedents application approach Artificial Intelligence BAGGER behavior BENCH3 block branching factor calculation causal model censor Chapter Computer Science concept Conference on Artificial constraints constructed deleted determine domain theory EBL system effects EGGS eralization example experiments explanation structure explanation-based learning expressions fact failure focus rule G. F. DeJong GENESIS goal Gripper heuristics Horn clauses inference rules initial instantiations International Joint Conference involves knowledge learning systems Machine Learning macro-operator moved narrative Natural Language Processing non-EBL object observed operator sequence osmosis partially ordered Peg1 performance physical joint schema PHYSICS 101 preconditions Principia problem problem-solving Proceedings produced proof R. J. Mooney real world requires robot rote rote-learning schemata sEBL shown in Figure simplifications situation situation calculus solution Solution1 solver solving specific std-EBL strategy supports T. M. Mitchell theorem tion tower understanding variables Wall4 Washer1 widget