Abstract
There are two major motivations for rule extraction from trained artificial neural networks. First, some of the proposed neural network architectures, like multiplayer perceptrons, are so complex that that it is difficult to understand the logic behind any decision or inference made by such a network. So from an engineering standpoint, rule extraction from such a complex network provides a way to understand and explain the logic behind any decision made by it. By the way, [11] define the rule extraction from neural networks task as follows: “Given a trained neural network and the examples used to train it, produce a concise and accurate symbolic description of the network.” So the objective of rule extraction is to provide a certain type of symbolic description of the network. A second major motivation for rule extraction is to bridge the divide between symbolic AI and connectionism; that is, to show that connectionist subsymbolic systems are just an implementation of higher-level symbolic systems. Thus rule-extraction and rule-insertion, whereby a connectionist network is created from a set of symbolic rules [16,17], provides a seamless integration between these two levels, the symbolic and the subsymbolic.
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
Abe, S. and M.-S. Lan (1995). Fuzzy Rule Extraction Directly from Numerical Data for Function Approximation. IEEE Trans. Systems, Man & Cybernetics, 25:119–129.
Alexander, J. A. & Mozer, M. C. (1995). Template-based algorithms for connectionist rule extraction. In Tesauro, G., Touretzky, D., & Leen, T., editors, Advances in Neural Information Processing Systems (volume 7). MIT Press.
Andrews, R., Diederich, J. and Tickle, A. B. (1995). A survey and critique of techniques for extracting rules from trained artificial neural networks. Knowledge-based Systems, 8, 6, 373–389.
Andrews, R. and J. Diederich, eds, (1996). Proceedings of the NIPS’96 Workshop on Rule Extraction From Trained Artificial Neural Networks. NIPS Foundation.
Apolloni, B., Malchiodi, D., Orovas, C., and Palmas, G. (2000a). Learning rule representations from data. Under review.
Apolloni, B., Malchiodi, D., Orovas, C., and Palmas, G. (2000b). From synapses to rules. In Foundations of Connectionist-symbolic Integration: Representation, Paradigms, and Algorithms — Proceedings of the 14th European Conference on Artificial Intelligence.
Boden, M. (1994). Horses of a Different Color? In: Honavar, V. and Uhr, L. (Ed.) Artificial Intelligence and Neural Networks: Steps Toward Principled Integration. New York: Academic Press.
Buckley, J. J. and Hayashi, Y. (1994). Fuzzy neural networks. In: Yager, R. and Zadeh, L. (Ed.) Fuzzy Sets, Neural Networks and Soft Computing. New York: Van Nostrand Reinhold.
Carpenter, G.A. and Tan, A.-H. (1995). Rule extraction: From neural architecture to symbolic representation. Connection Science, 7, 3–27.
Carpenter, G. A. and Grossberg, S. (1994). Fuzzy ARTMAP: A synthesis of neural networks and fuzzy logic for supervised categorization and nonstationary prediction. In: Yager, R. and Zadeh, L. (Ed.) Fuzzy Sets, Neural Networks and Soft Computing. New York: Van Nostrand Reinhold.
Craven, M. W. and Shavlik, J. W. (1994). Using sampling and queries to extract rules from trained neural networks. Machine Learning: Proceedings of the Eleventh International Conference, San Francisco, CA.
Fahlman, S. E. and Hinton, G. E. (1987). Connectionists Architectures for Artificial Intelligence. Computer, 20, 100–109.
Feldman, J. A. and Ballard, D. A. (1982). Connectionists Models and Their Properties. Cognitive Science, 6, 205–254.
Frasconi, P., Gori, M., Maggini, M., and Soda, G. (1995). Unified integration of explicit rules and learning by example in recurrent networks. IEEE Transactions on Knowledge and Data Engineering, vol. 7, no. 2, pp. 340–346.
Fu, L. M. (1994). Rule generation from neural networks. IEEE Transactions on Systems, Man and Cybernetics, 24 (8), pp. 1114–1124.
Giles, L. and C.W. Omlin (1993). Extraction, insertion and refinement of symbolic rules in dynamically-driven recurrent neural networks. Connection Science, 5(3,4):307–337, Special Issue on Architectures for Integrating Symbolic and Neural Processes.
Giles, L. and Christian W. Omlin (1994). Extraction and insertion of symbolic information in recurrent neural networks. In V. Honavar and L. Uhr, editors, Artificial Intelligence and Neural Networks: Steps toward Principled Integration, pages 271–299. Academic Press.
Goonatilake, S. and S. Khebbal (Ed.), (1995), Intelligent Hybrid Systems. New York: Wiley.
Grossberg, S. (1982). Studies of Mind and Brain: Neural Principles of Learning Perception, Development, Cognition, and Motor Control. Boston: Reidell Press.
Grossberg, S. (1987). Competitive learning: From interactive activation to adaptive resonance. Cognitive Science, 11, 23–63.
Grossberg, S. (1988). Nonlinear neural networks: principles, mechanisms, and architectures. Neural Networks, 1, 17–61.
Gupta, M. M. and Rao, D. H. (1994). On the principles of fuzzy neural networks. Fuzzy Sets and Systems, 61, 1, 1–18.
Haugeland, J. (1996). What is Mind Design. Chapter 1 in Haugeland, J. (ed), Mind Design II, 1997, MIT Press, 1–28.
Honavar, V. and Uhr, L. (Ed.), (1994), Artificial Intelligence and Neural Networks: Steps Toward Principled Integration. New York, NY: Academic Press.
Honavar, V. and Uhr, L. (1995). Integrating Symbol Processing Systems and Connectionist Networks. In: Goonatilake, S. and Khebbal, S. (Ed.) Intelligent Hybrid Systems. New York: Wiley.
Horwitz, B., Friston, K. J., and Taylor, J. G. (2000). Neural modeling and functional brain imaging: an overview. Neural Networks, Vol. 13, No. 8–9, 829–846.
Keller, J. M. and Hunt, D. (1985). Incorporating fuzzy membership functions into the perceptron algorithm. IEEE Trans. Pattern Anal. Machine Intelligence, 7, 693–699.
Kohonen, T. (1988). An introduction to neural networks. Neural Networks, 1, 3–16.
Kohonen, T. (1989). Self-organization and associative memory. 3rd ed. Berlin, Heidelberg: Spriger-Verlag.
Kosko, B. (1992). Neural Networks and Fuzzy Systems. Prentice Hall, Englewood Cliffs, NJ.
Levine, D. and Apariciov, M.. (Ed.) (1994). Neural Networks for Knowledge Representation. New York: Lawrence Erlbaum.
Lin, C. T. and Lee, C. S. G. (1994). Supervised and unsupervised learning with fuzzy similarity for neural network-based fuzzy logic control systems. In: Yager, R. and Zadeh, L. (Ed.) Fuzzy Sets, Neural Networks and Soft Computing. New York: Van Nostrand Reinhold.
McClelland, J. L. (1985). Putting knowledge in its place: A scheme for programming parallel processing structures on the fly. Cognitive Science,9,1130–146.
McClelland, J.L., McNaughton, B.L., and O’Reilly, R.C. (1995). Why there are complementary learning systems in hippocampus and neocortex: Insights from the successes and failures of connectionist models of learning and memory. Psychological Review. 102: 419–457.
Moody, J. & Darken, C. (1989). Fast Learning in Networks of Locally-Tuned Processing Units, Neural Computation, 1(2), 281–294.
Pal, S. K. & Mitra, S. (1992). Multi-layer perceptrons, fuzzy sets and classification, IEEE Transactions on Neural Networks, NN-3, 683–697.
Reilly, D.L., Cooper, L.N. and Elbaum, C. (1982). A Neural Model for Category Learning. Biological Cybernetics, 45, 35–41.
Roy, A. (2000). On Connectionism, Rule Extraction and Brain-like Learning. IEEE Transactions on Fuzzy Systems, Vol. 8, No. 2, pp. 222–227.
Roy, A., Kim, L.S. & Mukhopadhyay, S. (1993). A Polynomial Time Algorithm for the Construction and Training of a Class of Multilayer Perceptrons. Neural Networks, Vol. 6, No. 4, pp. 535–545.
Roy, A., Govil, S. & Miranda, R. (1995). An Algorithm to Generate Radial Basis Function (RBF)-like Nets for Classification Problems. Neural Networks, Vol. 8, No. 2, pp. 179–202.
Roy, A., Govil, S. & Miranda, R. (1997a). A Neural Network Learning Theory and a polynomial Time RBF Algorithm. IEEE Transactions on Neural Networks,8, 6, pp. 1301–1313.
Roy, A. & Mukhopadhyay, S. (1997b). Iterative Generation of Higher-Order Nets in Polynomial Time Using Linear Programming. IEEE Transactions on Neural Networks, 8, 2,402–412.
Roy, A. (1999). Brain’s internal mechanisms — a new paradigm. Proceedings of the International Joint Conference on Neural Networks (IJCNN’99), Washington, D.C., paper no. 259.
Rumelhart, D.E., and McClelland, J.L. (eds.)(1986). Parallel Distributed Processing: Explorations in Microstructure of Cognition, Vol. 1: Foundations. MIT Press, Cambridge, MA., 318–362.
Rumelhart, D.E. (1989). The Architecture of Mind: A Connectionist Approach. Chapter 8 in Haugeland, J. (ed). Mind Design II, 1997, MIT Press, 205–232.
Shadmehr, R. and Holcomb, H. (1997). Neural correlates of motor memory consolidation. Science, Vol. 277, pp. 821–825.
Setiono, R. and Liu, H. (1996). Symbolic Representation of Neural Networks. IEEE Computer, 29(3), pp. 71–76.
Setiono, R. and Liu, H. (1997). NeuroLinear: From neural networks to oblique decision rules. Neurocomputing, 17, pp. 1–24.
Simpson, P. K. (1992). Fuzzy min-max neural networks — part I: Classification. IEEE Trans. On Neural Networks, 3, 5, 776–786.
Smolensky, P. (1989). Connectionist Modeling: Neural Computation/Mental Connections. Chapter 9 in Haugeland, J. (ed), Mind Design II, 1997, MIT Press, 233–250.
Sun, R. (1994). Logic and Variables in Connectionist Networks: A Brief Overview. In: Honavar, V. and Uhr, L. (Ed.) Artificial Intelligence and Neural Networks: Steps Toward Principled Integration. New York: Academic Press.
Sun, R. (1994). Integrating Rules and Connectionism for Robust Commonsense Reasoning. John Wiley and Sons, New York, NY.
Sun, R. and Bookman, L. (Ed.) (1995), Computational Architectures Integrating Symbolic and Neural Processes. New York: Kluwer.
Takagi, H. & Hayasi, A. (1991). NN-driven fuzzy reasoning. Int. J. Approximate Reasoning. 5, 191–212.
Towell, G. and Shavlik, J. (1993). The extraction of refined rules from knowledge-based neural networks. Machine Learning, 13, 1, 71–101.
Yuan, F., Feldkamp, L.A., Davis, L.I., & Puskorius, G.V. (1992). Training a hybrid neural-fuzzy system. Proceedings of IJCNN’92, Baltimore, Vol. II, pp. 739–744.
Yager, R. R. (1993). Generalized fuzzy and matrix associative holographic memories. J. of Int. and Fuzzy Systems, 1, 1, 43–53.
Zadeh, L. A. (1965). Fuzzy Sets. Information and Control, 8, 338–353.
Zadeh, L. A. (1973). Outline of a new approach to the analysis of complex systems and decision process. IEEE Trans. Systems, Man and Cybernetics, 3, 1, 28–44.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag London Limited
About this paper
Cite this paper
Roy, A. (2002). On connectionism and rule extraction. In: Tagliaferri, R., Marinaro, M. (eds) Neural Nets WIRN Vietri-01. Perspectives in Neural Computing. Springer, London. https://doi.org/10.1007/978-1-4471-0219-9_31
Download citation
DOI: https://doi.org/10.1007/978-1-4471-0219-9_31
Publisher Name: Springer, London
Print ISBN: 978-1-85233-505-2
Online ISBN: 978-1-4471-0219-9
eBook Packages: Springer Book Archive