Machine Learning Proceedings 1991

Part I Automated Knowledge Acquisition Design Rationale Capture as Knowledge Acquisition A Domain-Independent Framework for Effective Experimentation in Planning Knowledge Refinement Using a High Level, Non-Technical Vocabulary Improving the Performance of Inconsistent Knowledge Bases Via Combined Optimization Method The Flexibility of Speculative Refinement Generating Error Candidates for Assigning Blame in a Knowledge BasePart II Computational Models of Human Learning A Prototype Based Symbolic Concept Learning System Combining Evidence of Deep and Surface Similarity The Importance of Causal Structure and Facts in Evaluating Explanations Learning Words From Context Modeling the Acquisition and Improvement of Motor Skills A Computational Model of Acquisition for Children's Addition Strategies Internal World Models and Supervised Learning Babel: A Psychologically Plausible Cross-Linguistic Model of Lexical and Syntactic Acquisition The Acquisition of Human Planning Expertise Adaptive Pattern-Oriented Chess Variability Bias and Category Learning A Constraint-Motivated Model of Lexical Acquisition Computer Modeling of Acquisition Orders in Child Language Simulating Stages of Human Cognitive Development With Connectionist Models Learning Physics Via Explanation-Based Learning of Correctness and Analogical Search ControlPart III Constructive Induction Incremental Constructive Induction: An Instance-Based Approach A Transformational Approach to Constructive Induction Learning Variable Descriptors for Applying Heuristics Across CSP Problems Informed Pruning in Constructive Induction A Hybrid Method for Feature Generation Abstracting Concepts with Inverse Resolution Opportunistic Constructive Induction Quantifying the Value of Constructive Induction, Knowledge, and Noise Filtering on Inductive Learning Discovering Production Rules with Higher Order Neural Networks Constructive Induction on Symbolic Features Comparison of Methods Based on Inverse Resolution The Need for Constructive Induction Constructive Induction in Theory Refinement Constructive Induction of M-of-N Terms Relations, Knowledge and Empirical Learning Learning Concepts by Synthesizing Minimal Threshold Gate Networks On the Effect of Instance Representation on Generalization Relational Clichés: Constraining Constructive Induction During Relational Learning Learning Polynomial Functions by Feature Construction Constructive Induction in Knowledge-Based Neural Networks Feature Construction in Structural Decision Trees Fringe-Like Feature Construction: A Comparative Study and a Unifying Scheme A Neural Network Approach to Constructive InductionPart IV Learning in Intelligent Information Retrieval Learning in Intelligent Information Retrieval A Probabilistic Retrieval Scheme for Cluster-Based Adaptive Information Retrieval Classification Trees for Information Retrieval Query Formulation Through Knowledge Acquisition Incremental Learning in a Probabilistic Information Retrieval System Query Learning Using an ANN with Adaptive Architecture A Goal-Based Approach to Intelligent Information Retrieval Machine Learning in the Combination of Expert Opinion Approach to IR Predicting Actions from Induction on Past PerformancePart V Learning Reaction Strategies Decision-Theoretic Learning in an Action System On Becoming Decreasingly Reactive: Learning to Deliberate Minimally Learning the Persistence of Actions in Reactive Control Rules Learning to Avoid Obstacles Through Reinforcement Learning Footfall Evaluation for a Walking Robot The Blind Leading the Blind: Mutual Refinement of Approximate Theories Learning to Select a Model in a Changing World Learning from Deliberated Reactivity Self-Improvement Based on Reinforcement Learning, Plannin