+ Site Statistics
+ Search Articles
+ PDF Full Text Service
How our service works
Request PDF Full Text
+ Follow Us
Follow on Facebook
Follow on Twitter
Follow on LinkedIn
+ Subscribe to Site Feeds
Most Shared
PDF Full Text
+ Translate
+ Recently Requested

Computation of madalines' sensitivity to input and weight perturbations

Computation of madalines' sensitivity to input and weight perturbations

Neural Computation 18(11): 2854-2877

The sensitivity of a neural network's output to its input and weight perturbations is an important measure for evaluating the network's performance. In this letter, we propose an approach to quantify the sensitivity of Madalines. The sensitivity is defined as the probability of output deviation due to input and weight perturbations with respect to overall input patterns. Based on the structural characteristics of Madalines, a bottom-up strategy is followed, along which the sensitivity of single neurons, that is, Adalines, is considered first and then the sensitivity of the entire Madaline network. By means of probability theory, an analytical formula is derived for the calculation of Adalines' sensitivity, and an algorithm is designed for the computation of Madalines' sensitivity. Computer simulations are run to verify the effectiveness of the formula and algorithm. The simulation results are in good agreement with the theoretical results.

Please choose payment method:

(PDF emailed within 0-6 h: $19.90)

Accession: 048617246

Download citation: RISBibTeXText

PMID: 16999581

DOI: 10.1162/neco.2006.18.11.2854

Related references

Sensitivity analysis of multilayer perceptron to input and weight perturbations. IEEE Transactions on Neural Networks 12(6): 1358-1366, 2001

The selection of weight accuracies for Madalines. IEEE Transactions on Neural Networks 6(2): 432-445, 1995

Temporal sensitivity of aphanizomenon flos aquae dominance a whole lake simulation study with input perturbations. Ecological Modelling 43(3-4): 137-154, 1988

Computation of Adalines' sensitivity to weight perturbation. IEEE Transactions on Neural Networks 17(2): 515-519, 2006

Real-time computing without stable states: a new framework for neural computation based on perturbations. Neural Computation 14(11): 2531-2560, 2002

Computation of trunk stability in forward perturbations: effects of preload, perturbation load, initial flexion and abdominal preactivation. Journal of Biomechanics 48(4): 716-720, 2015

Computation of the input impedances of a catheter for cardiac volumetry. IEEE Transactions on Bio-Medical Engineering 31(6): 448-453, 1984

Computation of equilibrium constants for the association of 2 different monomer units from weight average molecular weight data. Biopolymers 9(12): 1465-1474, 1970

Computation of molecular weight and weight fraction of five and six components in mixtures from model equilibrium ultracentrifugation data. Journal of Agricultural & Food Chemistry 40(5): 824-829, 1992

Response of marine aerosol to changes in phytoplankton induced by perturbations to aeolian iron input. Geochimica et Cosmochimica Acta 74.12, Suppl. 1: A1141, 2010

The computation of total sediment transport in view of changed input parameters. Pages 548-553:, 1989

Two- and three-input TALE-based AND logic computation in embryonic stem cells. Nucleic Acids Research 41(21): 9967-9975, 2013

A Role for Synaptic Input Distribution in a Dendritic Computation of Motion Direction in the Retina. Neuron 89(6): 1317-1330, 2016

Computation of reduced energy input current stimuli for neuron phase models. Conference Proceedings 2014: 4847-4851, 2014

The responses of simultaneously recorded purkinje cells to perturbations of the step cycle and their relationship to the climbing fiber input. Society for Neuroscience Abstracts 11(2): 1035, 1985