A cascading structure and training method for multilayer neural network

Li, Yan and Rad, A. B. (1997) A cascading structure and training method for multilayer neural network. International Journal of Neural Systems, 8 (5/6). pp. 509-515. ISSN 0129-0657

[img]
Preview
Text
Yan Rad 1997 Cascading NN Final.pdf

Download (523Kb) | Preview

Abstract

A new structure and training method for multilayer neural networks is presented. The proposed method is based on cascade training of subnetworks and optimizing weights layer by layer. The training procedure is completed in two steps. First, a subnetwork, m inputs and n outputs as the style of training samples, is trained using the training samples. Secondly the outputs of the subnetwork is taken as the inputs and the outputs of the training sample as the desired outputs, another subnetwork with n inputs and n outputs is trained. Finally the two trained subnetworks are connected and a trained multilayer neural networks is created. The numerical simulation results based on both linear least squares back-propagation (LSB) and traditional back-propagation (BP) algorithm have demonstrated the efficiency of the proposed method.


Statistics for USQ ePrint 25060
Statistics for this ePrint Item
Item Type: Article (Commonwealth Reporting Category C)
Refereed: Yes
Item Status: Live Archive
Additional Information: Access to Published Version restricted in accordance with publisher copyright policy.
Faculty / Department / School: Historic - Faculty of Engineering and Surveying - Department of Electrical, Electronic and Computer Engineering
Date Deposited: 15 Jul 2014 04:31
Last Modified: 05 Aug 2014 00:39
Uncontrolled Keywords: back-propagation algorithm; linear least squares back-propagation; neural networks
Fields of Research : 08 Information and Computing Sciences > 0801 Artificial Intelligence and Image Processing > 080108 Neural, Evolutionary and Fuzzy Computation
Socio-Economic Objective: E Expanding Knowledge > 97 Expanding Knowledge > 970108 Expanding Knowledge in the Information and Computing Sciences
Identification Number or DOI: 10.1142/S0129065797000495
URI: http://eprints.usq.edu.au/id/eprint/25060

Actions (login required)

View Item Archive Repository Staff Only