Creating A Memory of Causal Relationships: An Integration of Empirical and Explanation-based Learning Methods

by Michael J. Pazzani

Overview

This book presents a theory of learning new causal relationships by making use of perceived regularities in the environment, general knowledge of causality, and existing causal knowledge. Integrating ideas from the psychology of causation and machine learning, the author introduces a new learning procedure called theory-driven learning that uses abstract knowledge of causality to guide the induction process.

Known as OCCAM, the system uses theory-driven learning when new experiences conform to common patterns of causal relationships, empirical learning to learn from novel experiences, and explanation-based learning when there is sufficient existing knowledge to explain why a new outcome occurred. Together these learning methods construct a hierarchical organized memory of causal relationships. As such, OCCAM is the first learning system with the ability to acquire, via empirical learning, the background knowledge required for explanation-based learning.

Please note: This program runs on common lisp.

Product Details

ISBN-13:	9781134992324
Publisher:	Taylor & Francis
Publication date:	03/07/2013
Sold by:	Barnes & Noble
Format:	eBook
Pages:	360
File size:	13 MB
Note:	This product may take a few minutes to download.

Table of Contents

Contents: Introduction. What OCCAM Is Up Against. Similarity-Based Learning in OCCAM. Theory-Driven Learning in OCCAM. Explanation-Based Learning in OCCAM. Integration of Learning Methods. Experiments in Integrated Learning. Future Directions and Conclusions. Appendices: Data Listing. Program Traces. Prolog OCCAM. OCCAM's Generalization Rules. Listing of Economic Sanction Incidents.

Editorial Reviews

"...the best source for learning about OCCAM...a clear and detailed description of OCCAM ....incorporates many interesting and novel ideas. The ideas that are emphasized by Pazzani, and that are explored in depth in the book, are a novel integration of explanation-based and similarity-based learning methods..."
—Machine Learning

From the Publisher