Stability of Back Propagation Training Algorithm for Neural Networks

Authors

Luma. N. M. Tawfiq

Abstract

In this paper, we derive and prove the stability bounds of the momentum coefficient µ and the learning rate η of the back propagation updating rule in Artificial Neural Networks .The theoretical upper bound of learning rate η is derived and its practical approximation is obtained

Keywords

BP training algorithm, Ann, Gradient Descent Training Algorithm, Gradient Descent with Momentum and Lyapunov Stability

Article Type

Article

How to Cite this Article

Tawfiq, Luma. N. M. (2012) "Stability of Back Propagation Training Algorithm for Neural Networks," Baghdad Science Journal: Vol. 9: Iss. 4, Article 19.
DOI: https://doi.org/10.21123/bsj.2012.9.4.713-719