Extracting rules from trained neural networks
- 1 March 2000
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks
- Vol. 11 (2) , 377-389
- https://doi.org/10.1109/72.839008
Abstract
This paper presents an algorithm for extracting rules from trained neural networks. The algorithm is a decompositional approach which can be applied to any neural network whose output function is monotone such as sigmoid function. Therefore, the algorithm can be applied to multilayer neural networks, recurrent neural networks and so on. It does not depend on training algorithms, and its computational complexity is polynomial. The basic idea is that the units of neural networks are approximated by Boolean functions. But the computational complexity of the approximation is exponential, and so a polynomial algorithm is presented. The author has applied the algorithm to several problems to extract understandable and accurate rules. This paper shows the results for the votes data, mushroom data, and others. The algorithm is extended to the continuous domain, where extracted rules are continuous Boolean functions. Roughly speaking, the representation by continuous Boolean functions means the representation using conjunction, disjunction, direct proportion, and reverse proportion. This paper shows the results for iris data.Keywords
This publication has 11 references indexed in Scilit:
- Survey and critique of techniques for extracting rules from trained artificial neural networksPublished by Elsevier ,2000
- Structural learning with forgettingNeural Networks, 1996
- Extraction of rules from discrete-time recurrent neural networksNeural Networks, 1996
- Combining symbolic and neural learningMachine Learning, 1994
- Using Sampling and Queries to Extract Rules from Trained Neural NetworksPublished by Elsevier ,1994
- Rule generation from neural networksIEEE Transactions on Systems, Man, and Cybernetics, 1994
- Extracting refined rules from knowledge-based neural networksMachine Learning, 1993
- Constant depth circuits, Fourier transform, and learnabilityJournal of the ACM, 1993
- Learning Symbolic Rules Using Artificial Neural NetworksPublished by Elsevier ,1993
- Self-learning fuzzy controllers based on temporal backpropagationIEEE Transactions on Neural Networks, 1992