Learning from good and bad data

Philip D. Laird9780898382631, 0-89838-263-7

Learning from Good and Bad Data explains the firm theoretical foundation that underlies much of the experimental research in machine learning. While the thrust of the work is theoretical, the presentation is accessible to theorists and practitioners, specialists and nonspecialists in the rapidly developing field of machine learning. Empirical learning (learning from example) is studied mathematically in order to uncover the formal structures common to much of the artificial intelligence experimental work on the subject.