Vectorized Backpropagation and Automatic Pruning for MLP Network Optimization

Stalin, Suryan and Sreenivas, TV (1993) Vectorized Backpropagation and Automatic Pruning for MLP Network Optimization. In: IEEE International Conference on Neural Networks, 1993, April, San Francisco, CA, vol.3, 1427-1432.

Preview

PDF Vectorized.pdf Download (491kB)

Abstract

In complicated tasks such as speech recognition, neural network architectures have to be improved for better learning and recognition performance. This paper presents an analysis of the backpropagation algorithm and reveals the significance of vectorized backpropagation and automatic pruning for better learning performance and MLP network optimization. During the learning phase, the network which uses vectorized backpropagation converges within 20% - 50% of the iterations required for the standard MLP to converge without affecting the test set performance. The network pruning algorithm reduces the number of hidden nodes and connection weights. The pruned network withonly 40% connection weights of the unpruned network gives the same learning and recognition performance as the parent unpruned fully connected network.

Item Type:	Conference Paper
Publisher:	IEEE
Additional Information:	Copyright 1990 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
Department/Centre:	Division of Electrical Sciences > Electrical Communication Engineering
Date Deposited:	25 Aug 2008
Last Modified:	19 Sep 2010 04:27
URI:	http://eprints.iisc.ac.in/id/eprint/6890

Actions (login required)

View Item