Abstract
Explanation is an important function in symbolic artificial intelligence (AI). For example, explanation is used in machine learning and for the interpretation of prediction failures in case-based reasoning. Furthermore, the explanation of results of a reasoning process to a user who is not a domain expert must be a component of any inference system. Experience with expert systems has shown that the ability to generate explanations is absolutely crucial for the user-acceptance of AI systems (Davis, Buchanan & Shortliffe 1977). In contrast to symbolic systems, neural networks have no explicit, declarative knowledge representation and therefore have considerable difficulties in generating explanation structures. In neural networks, knowledge is encoded in numeric parameters (weights) and distributed all over the system.
It is the intention of this paper ro discuss the ability of connectionist systems to generate explanations. It will be shown that connectionist systems benefit from the explicit encoding of relations and the use of highly structured networks in order to realize explanation and explanation components. Furthermore, structured connectionist systems using spreading activation have the advantage that any intermediate state in processing is semantically meaningfull and cam be used for explanation. The paper describes sseveral successful applications of explanation possible future realizations of explanation in neural networks.
Research reported in this paper was done for the most part while the author was affiliated with the International Computer Science Institute in Berkeley, California.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Charniak, E. (1987): Connectionism and Explanation. Paper presented at TINLAP-3, New Mexico State University, Las Cruses, January 1987.
Davis, R., Buchanan, B.G. & Shortliffe, E. (1977): Production Rules as a Representation for a Knowl¬edge-Based Consultation Program. Artificial Intelligence 8 (1), 15 - 45.
Diederich, J. (1988a): Connectionist Recruitment Learning. ECAI-88, Proc. of the 8th European Confer-ence on Artificial Intelligence, Munich, 351 - 356.
Diederich, J. (1988b): Knowledge-Intensive Recruitment Learning. ICSI-TR-88-010, International Com-puter Science Institute, Berkeley, November 1988.
Diederich, J. (1989a): Steps toward knowledge-intensive connectionist learning. In: Pollack, J. & Barnden, J. (Eds.): Advances in Connectionist and Neural Computation Theory. Ablex Publ.
Diederich, J. (1989b): “Learning by instruction” in connectionist systems. Proc. of the 6th International Workshop on Machine Learning. Cornell University, June/July 1989
Feldman, J. A. (1989): personal communication.
Hinton, G.E. (1988): personal communication.
Goel, A., Ramanujam, J. & Sadayappan, P. (1988): Towards a neural architecture for abductive reasoning. Proceedings of the 2nd IEEE International Conference on Neural Networks. Vol. 1, 681 - 688.
Lewis, C. (1988): Why and How to Learn Why: Analysis-based Generalization of Procedures. Cognitive Science, 12, 211-256.
Lischka, C. (1989): Konnektionismus, Künstliche Intelligenz und Informatik - Einige Anmerkungen. Sub-mitted to: “KI.”
Mitchell, T.M., Keller, R.M. & Kedar-Cabelli, S.T. (1986): Explanation-Based Generalization: A Unifying View. Machine Learning 1, 47-80.
Peng, Y. & Reggia, J.A. (1989): A Connectionist Model for Diagnostic Problem Solving. IEEE Transac-tions on Systems, Man and Cybernetics. In press.
Slade, S. (1988): Case-based reasoning: A Research Paradigm. YALEU/CSD/RR #644, Yale University, New Haven, Conn., August 1988.
Shastri, L. (1988): Semantic Networks: An Evidential Formalization and its Connectionist Realization. Morgan Kaufman Publ., San Mateo, California 1988.
Shastri, L. & Ajjanagadde, V. (1989): A Connectionist System for Rule Based Reasoning with Multi-Place Predicates and Variables. MS-CIS-8905 & LINC LAB 141, Computer and Information Science Depart-ment, University of Pennsylvania, January 1989
Thagard, P. (1988): Explanatory coherence. Final draft for commentators. Behavioral and Brain Sciences. Cambridge University Press. December 1988.
Wald, J., Farach, M., Tagamets, M. & Reggia, J.A. (1989): Generating Plausible Diagnostic Hypotheses with Self-Processing Causal Networks. Journal of Experimental and Theoretical Artificial Intelligence. In press.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1989 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Diederich, J. (1989). Explanation and Connectionism. In: Metzing, D. (eds) GWAI-89 13th German Workshop on Artificial Intelligence. Informatik-Fachberichte, vol 216. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-75100-4_14
Download citation
DOI: https://doi.org/10.1007/978-3-642-75100-4_14
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-51743-6
Online ISBN: 978-3-642-75100-4
eBook Packages: Springer Book Archive