Fault immunization technique for artificial neural networks

By injecting some chemical substances to a cell, it is able to enhance the ability of the cell to fight against the intruder. This immunization concept in biological cells has been applied to enhance the fault tolerance capability in a perceptron-like neuron. In this paper, we consider only the case where each neuron separates its input vectors into two classes. We mathematically model the cell immunization in terms of weight vector relocation and propose a polynomial time weight relocating algorithm. This algorithm can be generalized to the case where each neuron separating the input vectors into more than two classes.

[1]  Anders Krogh,et al.  Introduction to the theory of neural computation , 1994, The advanced book program.

[2]  Chun-Shin Lin,et al.  Maximizing fault tolerance in multilayer neural networks , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[3]  Benjamin W. Wah,et al.  Fault tolerant neural networks with hybrid redundancy , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[4]  R. Palmer,et al.  Introduction to the theory of neural computation , 1994, The advanced book program.

[5]  Robert Chun,et al.  Immunization of neural networks against hardware faults , 1990, IEEE International Symposium on Circuits and Systems.

[6]  William Fornaciari,et al.  A new architecture for the automatic design of custom digital neural network , 1995, IEEE Trans. Very Large Scale Integr. Syst..

[7]  R.D. Clay,et al.  Limiting fault-induced output errors in ANNs , 1991, IJCNN-91-Seattle International Joint Conference on Neural Networks.

[8]  Chidchanok Lursinsap,et al.  Weight shifting techniques for self-recovery neural networks , 1994, IEEE Trans. Neural Networks.