Theoretical investigation of the robustness of multilayer perceptrons: analysis of the linear case and extension to nonlinear networks

Authors:
P. Kerlirzin;P. Refregier
Affiliations:
Central Res. Lab., Thomson-CSF, Orsay;-
Venue:
IEEE Transactions on Neural Networks
Year:
1995

Citing 0
Cited 3

A low-sensitivity recurrent neural network

Neural Computation
Some neural network applications in environmental sciences. Part I: forward and inverse problems in geophysical remote measurements

Neural Networks - 2003 Special issue: Neural network analysis of complex scientific data: Astronomy and geosciences
Neural Networks and Metabolic Networks: Fault Tolerance and Robustness Features

Proceedings of the 2009 conference on Neural Nets WIRN09: Proceedings of the 19th Italian Workshop on Neural Nets, Vietri sul Mare, Salerno, Italy, May 28--30 2009

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we address the problem of robustness in multilayer perceptrons. We present the main theoretical results in the case of linear neural networks with one hidden layer in order to go beyond the empirical study we previously made. We show that the robustness can greatly be improved and that even without decreasing performance in normal use. Finally, we show how this behavior, clearly demonstrated in the linear case, is an approximation of the behavior of nonlinear networks