Extending Explanation-based Learning by Generalizing the Structure of ExplanationsFor researchers in machine learning, describes three systems for generalizing explanations, the PHYSICS 101 system, the BAGGER system, and its successor, BAGGER2. Presents details of their algorithms, and discusses several example of learning by each. No index. Annotation copyright Book News, Inc. |
Contents
Learning in MathematicallyBased Domains | 15 |
A DomainIndependent Approach | 71 |
An Empirical Analysis of ExplanationBased Learning | 117 |
Copyright | |
5 other sections not shown
Other editions - View all
Common terms and phrases
acquired rules additional analyzing approach arbitrary number Artificial Intelligence autonomous mode BAGGER system block blockers blocks world calculation cancellation graph clear collection COMMON LISP concepts consequent constraints construct contains DeMorgan's Law determine disjuncts domain theory efficiency eliminated equation schemata experiments explanation structure expression external forces Fext Figure final Fint Fnet focus rule formulae goal implementation inference rules instantiation Intelligent Tutoring Systems inter-object forces INTERLISP intermediate involves knowledge learning systems left-hand side Machine Learning mathematical momentum moved NO-LEARN node object ObjectsInWorld obstacleSet operator performance PHYSICS 101 preconditions predicate presents primary obstacles problem solver Problem-Solving Schema produces proof recurrence recursive requires rule applications rules learned sample problem satisfy secondary obstacles sequence sequential rule situation calculus solving special-case specific example specific problem step strategy structure of explanations subexplanations table2 technique tower-building training mode unwindable rules variables vector