Limiting the effects of weight errors in feedforward networks using interval arithmetic

We address in this work the problem of weight inaccuracies in digital and analog feedforward networks. Both kind of implementations suffer from this problem due to physical limits of the particular technology. This work presents a novel and effective approach through the application of interval arithmetic to the multilayer perceptron. Results show that our method allows one to (1) compute strict bounds of the output error of the network, (2) find robust solutions respect to weight inaccuracies and (3) compute the minimum weight precision required to obtain the desired performance of the network.

[1]  Steve W. Piche,et al.  The selection of weight accuracies for Madalines , 1995, IEEE Trans. Neural Networks.

[2]  Jorma Rissanen,et al.  Stochastic Complexity in Statistical Inquiry , 1989, World Scientific Series in Computer Science.

[3]  Hideo Tanaka,et al.  An extension of the BP-algorithm to interval input vectors-learning from numerical data and expert's knowledge , 1991, [Proceedings] 1991 IEEE International Joint Conference on Neural Networks.

[4]  Markus Höhfeld,et al.  Probabilistic rounding in neural network learning with limited precision , 1992, Neurocomputing.

[5]  Bernard Widrow,et al.  Sensitivity of feedforward neural networks to weight errors , 1990, IEEE Trans. Neural Networks.

[6]  Yun Xie,et al.  Analysis of the effects of quantization in multilayer neural networks using a statistical model , 1992, IEEE Trans. Neural Networks.

[7]  Hiroshi Yamamoto,et al.  Reduction of required precision bits for back-propagation applied to pattern recognition , 1993, IEEE Trans. Neural Networks.

[8]  H. Ishibuchi,et al.  An architecture of neural networks with interval weights and its application to fuzzy regression analysis , 1993 .

[9]  G. Alefeld,et al.  Introduction to Interval Computation , 1983 .

[10]  David E. Culler,et al.  A case for NOW (networks of workstation) , 1995, PODC '95.

[11]  Chong-Ho Choi,et al.  Sensitivity analysis of multilayer perceptron with differentiable activation functions , 1992, IEEE Trans. Neural Networks.

[12]  Y. Xie,et al.  Analysis of effects of quantisation in multilayer neural networks using statistical model , 1991 .

[13]  O. H. Lowry.  Academic press. , 1972, Analytical chemistry.

[14]  Davide Anguita,et al.  Incorporating a priori knowledge into neural networks , 1995 .

[15]  John Wawrzynek,et al.  Using simulations of reduced precision arithmetic to design a neuro-microprocessor , 1993, J. VLSI Signal Process..