Machine Learning Proceedings 1992

Machine Learning Proceedings 1992

Proceedings of the Ninth International Workshop (ML92)

1st Edition - June 1, 1992

Write a review

  • Editors: Peter Edwards, Derek Sleeman
  • eBook ISBN: 9781483298535

Purchase options

Purchase options
DRM-free (PDF)
Sales tax will be calculated at check-out

Institutional Subscription

Free Global Shipping
No minimum order


Machine Learning: Proceedings of the Ninth International Workshop (ML92) covers the papers and posters presented at ML92, the Ninth International Machine Learning Conference, held at Aberdeen, Scotland on July 1-3, 1992. The book focuses on the advancements of practices, methodologies, approaches, and techniques in machine learning. The selection first offers information on the principal axes method for constructive induction; learning by incomplete explanations of failures in recursive domains; and eliminating redundancy in explanation-based learning. Topics include means-ends analysis search in recursive domains, description space transformation, distance metric, generating similarity matrix, and learning principal axes. The text then examines trading off consistency and efficiency in version-space induction; improving path planning with learning; finding the conservation of momentum; and learning to predict in uncertain continuous tasks. The manuscript elaborates on a teaching method for reinforcement learning, compiling prior knowledge into an explicit bias, spatial analogy and subsumption, and multistrategy learning with introspective meta-explanations. The publication also ponders on selecting typical instances in instance-based learning and temporal difference learning of backgammon strategy. The selection is a valuable source of information for researchers interested in machine learning.

Table of Contents

  • Generalizing from Case Studies: A Case Study

    On Learning More Concepts

    The Principal Axes Method for Constructive Induction

    Learning by Incomplete Explanations of Failures in Recursive Domains

    Eliminating Redundancy in Explanation-Based Learning

    Trading Off Consistency and Efficiency in Version-Space Induction

    Peepholing: Choosing Attributes Efficiently for Megainduction

    Improving Path Planning with Learning

    The Right Representation for Discovery: Finding the Conservation of Momentum

    Learning to Predict in Uncertain Continuous Tasks

    Lazy Partial Evaluation: An Integration of Explanation-Based Generalisation and Partial Evaluation

    A Teaching Method for Reinforcement Learning

    Compiling Prior Knowledge into an Explicit Bias

    Spatial Analogy and Subsumption

    Learning to Satisfy Conjunctive Goals

    Multistrategy Learning with Introspective Meta-Explanations

    An Asymptotic Analysis of Speedup Learning

    Why EBL Produces Overly-Specific Knowledge: A Critique of the PRODIGY Approaches

    Automatic Feature Generation for Problem Solving Systems

    Towards Inductive Generalisation in Higher Order Logic

    Ordering Effects in Clustering

    Learning Structured Concepts Using Genetic Algorithms

    An Analysis of Learning to Plan as a Search Problem

    An Approach to Anytime Learning

    Artificial Universes—Towards a Systematic Approach to Evaluating Algorithms Which Learn from Examples

    Average Case Analysis of Learning k-CNF Concepts

    The MENTLE Approach to Learning Heuristics for the Control of Logic Programs

    Fuzzy Substructure Discovery

    Efficient Classification of Massive, Unsegmented Datastreams

    Induction of One-Level Decision Trees

    Combining Competition and Cooperation in Supervised Inductive Learning

    A Practical Approach to Feature Selection

    Learning as Optimization: Stochastic Generation of Multiple Knowledge

    Dynamic Optimization

    Sub-Unification: A Tool for Efficient Induction of Recursive Programs

    Augmenting and Efficiently Utilizing Domain Theory in Explanation-Based Natural Language Acquisition

    Enhancing Transfer in Reinforcement Learning by Building Stochastic Models of Robot Actions

    THOUGHT: An Integrated Learning System for Acquiring Knowledge Structure

    An Approach to Concept Learning Based on Term Generalization

    Using Transitional Proximity for Faster Reinforcement Learning

    NFDT: A System that Learns Flexible Concepts Based on Decision Trees for Numerical Attributes

    A Symbolic Algorithm for Computing Coefficients' Accuracy in Regression

    Compression, Significance, and Accuracy

    Guiding Example Acquisition by Generating Scenarios

    Constructive Induction Using a Non-Greedy Strategy for Feature Selection

    Training Second-Order Recurrent Neural Networks Using Hints

    DYNAMIC: A New Role for Training Problems in EBL

    A Framework for Discovering Discrete Event Models

    Learning Episodes for Optimization

    Learning to Fly

    Deconstructing the Digit Recognition Problem

    On Combining Multiple Speedup Techniques

    Scaling Reinforcement Learning Algorithms by Learning Variable Temporal Resolution Models

    Detecting Novel Classes with Applications to Fault Diagnosis

    Measuring Utility and the Design of Provably Good EBL Algorithms

    Refining a Relational Theory with Multiple Faults in the Concept and Subconcepts

    Cooperation in Knowledge Base Refinement

    Temporal Difference Learning of Backgammon Strategy

    AGIL: Solving the Exploration Versus Exploitation Dilemma in a Simple Classifier System Applied to Simulated Robotics

    Conceptual Clustering with Systematic Missing Values

    Selecting Typical Instances in Instance-Based Learning

    The First Phase of Real-World Discovery: Determining Repeatability and Error of Experiments

    Author Index

    Subject Index

Product details

  • No. of pages: 448
  • Language: English
  • Copyright: © Morgan Kaufmann 1992
  • Published: June 1, 1992
  • Imprint: Morgan Kaufmann
  • eBook ISBN: 9781483298535

About the Editors

Peter Edwards

Affiliations and Expertise

University of Glamorgan

Derek Sleeman

Ratings and Reviews

Write a review

There are currently no reviews for "Machine Learning Proceedings 1992"