Designing neural network explanation facilities using genetic algorithms

The authors describe the use of genetic algorithms to provide components of explanation facilities for neural network applications. The genetic algorithm implementation, Genesis, uses a trained backpropagation neural network weight matrix as the genetic algorithm fitness function. Using different combinations of Genesis' run-time options, codebook vectors and decision surfaces are defined for the trained neural network. These vectors and surfaces can then be used as components of a facility that explains how the network is trained, and how it differentiates between classes. Two examples of this methodology are presented and briefly discussed. The first is a network trained to solve the XOR problem. The second is a network trained to diagnose appendicitis.<<ETX>>

[1]  Russell C. Eberhart,et al.  Neural network paradigm comparisons for appendicitis diagnoses , 1991, [1991] Computer-Based Medical Systems@m_Proceedings of the Fourth Annual IEEE Symposium.

[2]  L. Darrell Whitley,et al.  Genetic Reinforcement Learning with Multilayer Neural Networks , 1991, ICGA.

[3]  Russell C. Eberhart,et al.  Neural Network Versus Bayesian Diagnosis Of Append:CITIS , 1990, [1990] Proceedings of the Twelfth Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[4]  Lawrence. Davis,et al.  Handbook Of Genetic Algorithms , 1990 .

[5]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[6]  L. Darrell Whitley,et al.  Genetic algorithms and neural networks: optimizing connections and connectivity , 1990, Parallel Comput..

[7]  John R. Koza,et al.  Genetic generation of both the weights and architecture for a neural network , 1991, IJCNN-91-Seattle International Joint Conference on Neural Networks.