Abstract
The paper argues for the usefulness of plausible explanations not just for analytical learning, but also for empirical generalization. The larger context is an implemented system that learns complex rules (for a musical task) on the basis of a qualitative theory of the domain. It learns by generalizing and compiling plausible explanations, but it can also incrementally modify learned rules in reaction to new evidence. The paper shows how this incremental modification (generalization) becomes more effective if it is based on an analysis of the explanations underlying learned rules; these explanations support a notion of ‘deep’ similarity and can provide substantial bias on the empirical modification of concepts. Several criteria that implement this bias are described, and an extended example illustrates how they lead to intelligent generalization behaviour.
Chapter PDF
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Collins, A. and Michalski, R. (1989). The Logic of Plausible Reasoning: A Core Theory. Cognitive Science 13, 1–49.
Danyluk, A. (1987). The Use of Explanations for Similarity-Based Learning. In Proceedings of the Tenth International Joint Conference on Artificial Intelligence (IJCAI-87), Milano, Italy. Morgan Kaufmann.
Danyluk, A. (1989). Finding New Rules for Incomplete Theories: Explicit Biases for Induction with Contextual Information. In Proceedings of the Sixth International Workshop on Machine Learning, Ithaca, N.Y. Morgan Kaufmann.
DeJong, G. (1989). Explanation-Based Learning with Plausible Inferencing. In Proceedings of the Fourth European Working Session on Learning (EWSL-89), Montpellier, France. Pitman Publishing.
Forbus, K.D. (1984). Qualitative Process Theory. Artificial Intelligence 24(1–3) (special volume on qualitative reasoning about physical systems), 85–169.
Mitchell, T., Mahadevan, S. and Steinberg, L. (1985). LEAP: A Learning Apprentice for VLSI Design. In Proceedings of the Ninth International Joint Conference on Artificial Intelligence (IJCAI-85), Los Angeles, CA. Morgan Kaufmann.
Mitchell, T., Keller, R. and Kedar-Cabelli, S. (1986). Explanation-Based Generalization: A Unifying View. Machine Learning 1(1).
Russell, S. (1986). Analogical and Inductive Reasoning. Ph.D. Thesis, Report STAN-CS-87-1150, Stanford University, Stanford, CA.
Tecuci, G. and Kodratoff, Y. (1989). Multi-Strategy Learning in Nonhomogeneous Domain Theories. In Proceedings of the Sixth International Workshop on Machine Learning, Ithaca, N.Y. Morgan Kaufmann.
Widmer, G. (1989). A Tight Integration of Deductive and Inductive Learning. In Proceedings of the Sixth International Workshop on Machine Learning, Ithaca, N.Y. Morgan Kaufmann.
Widmer, G. (1990). A Qualitative Model of the Perception of Melodies and Harmonizations. Submitted. Available as Report TR-90-12, Austrian Research Institute for Artificial Intelligence, Vienna.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1991 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Widmer, G. (1991). Using plausible explanations to bias empirical generalization in weak theory domains. In: Kodratoff, Y. (eds) Machine Learning — EWSL-91. EWSL 1991. Lecture Notes in Computer Science, vol 482. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0017002
Download citation
DOI: https://doi.org/10.1007/BFb0017002
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-53816-5
Online ISBN: 978-3-540-46308-5
eBook Packages: Springer Book Archive