Relationships Between the A Priori and A Posteriori Errors in Nonlinear Adaptive Neural Filters

Authors:
Danilo P. Mandic;Jonathon A. Chambers
Affiliations:
School of Information Systems, University of East Anglia, Norwich, U.K.;Signal Processing Section, Department of Electrical and Electronic Engineering, Imperial College of Science, Technology and Medicine, London, U.K.
Venue:
Neural Computation
Year:
2000

Citing 7
Cited 3

Theory and design of adaptive filters

Theory and design of adaptive filters
Adaptive filter theory (3rd ed.)

Adaptive filter theory (3rd ed.)
Relating the slope of the activation function and the learning rate within a recurrent neural network

Neural Computation
Optimization by Vector Space Methods

Optimization by Vector Space Methods
Neural Networks: A Comprehensive Foundation

Neural Networks: A Comprehensive Foundation
Global asymptotic convergence of nonlinear relaxation equations realised through a recurrent perceptron

ICASSP '99 Proceedings of the Acoustics, Speech, and Signal Processing, 1999. on 1999 IEEE International Conference - Volume 02
The interchangeability of learning rate and gain in backpropagation neural networks

Neural Computation

Data-reusing recurrent neural adaptive filters

Neural Computation
A Data-Reusing Nonlinear Gradient Descent Algorithm for a Class of Complex-Valued Neural Adaptive Filters

Neural Processing Letters
A homomorphic neural network for modeling and prediction

Neural Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

The lower bounds for the a posteriori prediction error of a nonlinear predictor realized as a neural network are provided. These are obtained for a priori adaptation and a posteriori error networks with sigmoid nonlinearities trained by gradient-descent learning algorithms. A contractivity condition is imposed on a nonlinear activation function of a neuron so that the a posteriori prediction error is smaller in magnitude than the corresponding a priori one. Furthermore, an upper bound is imposed on the learning rate eta so that the approach is feasible. The analysis is undertaken for both feedforward and recurrent nonlinear predictors realized as neural networks.