A Complement to the Back-Propagation Algorithm: An Upper Bound for the Learning Rate

Jés J.F. Cerqueira; Álvaro G.B. Palhares; Marconi K. Madrid

doi:10.1109/IJCNN.2000.860823

Previous
Next
Table of Contents

Neural Networks, IEEE - INNS - ENNS International Joint Conference on

A Complement to the Back-Propagation Algorithm: An Upper Bound for the Learning Rate

Year: 2000, Volume: 5, Pages: 4517

DOI Bookmark: 10.1109/IJCNN.2000.860823

Authors

Jés J.F. Cerqueira, Universidade Estadual de Campinas
Álvaro G.B. Palhares, Universidade Estadual de Campinas
Marconi K. Madrid, Universidade Estadual de Campinas

Abstract

This paper presents as complement for the back-propagation algorithm, a local convergence analysis gotten from a vectorial approach. The analysis is made through application of Lyapunov's second method, and it supplies an upper bound for the learning rate.

Like what you’re reading?

Already a member?

Get this article FREE with a new membership!