Regularization and complexity control in feed-forward networks

Bishop, Christopher M. (1995). Regularization and complexity control in feed-forward networks. IN: Proceedings International Conference on Artificial Neural Networks ICANN'95. EC2 et Cie. (Unpublished)

Abstract

In this paper we consider four alternative approaches to complexity control in feed-forward networks based respectively on architecture selection, regularization, early stopping, and training with noise. We show that there are close similarities between these approaches and we argue that, for most practical applications, the technique of regularization should be the method of choice.

Divisions: ?? 13770100JJ ??
Additional Information: International Conference on Artificial Neural Networks ICANN'95.
Uncontrolled Keywords: NCRG complexity control feed-forward networks architecture selection regularization early stopping training with noise
Published Date: 1995

Download

[img]

Export / Share Citation


Statistics

Additional statistics for this record