A Quantified Sensitivity Measure for Multilayer Perceptron to Input Perturbation
- 1 January 2003
- journal article
- Published by MIT Press in Neural Computation
- Vol. 15 (1), 183-212
- https://doi.org/10.1162/089976603321043757
Abstract
The sensitivity of a neural network's output to its input perturbation is an important issue with both theoretical and practical values. In this article, we propose an approach to quantify the sensitivity of the most popular and general feedforward network: multilayer perceptron (MLP). The sensitivity measure is defined as the mathematical expectation of output deviation due to expected input deviation with respect to overall input patterns in a continuous interval. Based on the structural characteristics of the MLP, a bottom-up approach is adopted. A single neuron is considered first, and algorithms with approximately derived analytical expressions that are functions of expected input deviation are given for the computation of its sensitivity. Then another algorithm is given to compute the sensitivity of the entire MLP network. Computer simulations are used to verify the derived theoretical formulas. The agreement between theoretical and experimental results is quite good. The sensitivity measure can be used to evaluate the MLP's performance.Keywords
This publication has 9 references indexed in Scilit:
- Using function approximation to analyze the sensitivity of MLP with antisymmetric squashing activation functionIEEE Transactions on Neural Networks, 2002
- Sensitivity analysis of multilayer perceptron to input and weight perturbationsIEEE Transactions on Neural Networks, 2001
- Sensitivity analysis of neocognitronIEEE Transactions on Systems, Man and Cybernetics, Part C (Applications and Reviews), 1999
- Perturbation method for deleting redundant inputs of perceptron networksNeurocomputing, 1997
- Sensitivity to errors in artificial neural networks: a behavioral approachIEEE Transactions on Circuits and Systems I: Regular Papers, 1995
- The selection of weight accuracies for MadalinesIEEE Transactions on Neural Networks, 1995
- Input Noise Immunity of Multilayer PerceptronsETRI Journal, 1994
- Sensitivity analysis of multilayer perceptron with differentiable activation functionsIEEE Transactions on Neural Networks, 1992
- Sensitivity of feedforward neural networks to weight errorsIEEE Transactions on Neural Networks, 1990