eCite Digital Repository

HONNs with ELM algorithm for medical applications

Citation

Xu, S and Liu, Y, HONNs with ELM algorithm for medical applications, Proceedings of the 12th International Conference on Control, Automation, Robotics and Vision, 5-7 December 2012, Guangzhou, China, pp. 1215-1219. ISBN 978-1-4673-1872-3 (2012) [Refereed Conference Paper]

Copyright Statement

COpyright 2012 IEEE

DOI: doi:10.1109/ICARCV.2012.6485360

Abstract

Higher Order Neural Networks (HONNs) are Artificial Neural Networks (ANNs) in which the net input to a computational neuron is a weighted sum of products of its inputs (rather than just a weighted sum of its inputs as in traditional ANNs). It was known that HONNs can implement invariant pattern recognition as well as handling high frequency and high order nonlinear business data. Extreme Learning Machine (ELM) randomly chooses hidden neurons and analytically determines the output weights. With ELM algorithm, only the connection weights between hidden layer and output layer are adjusted. This paper develops an ELM algorithm for HONN models and applies it in several significant medical cases. The experimental results demonstrate significant advantages of HONN models with ELM algorithm such as faster training and improved generalization abilities (in comparison with standard HONN models).

Item Details

Item Type:Refereed Conference Paper
Keywords:artificial neural network, feedforward neural network, higher order neural network, extreme learning machine
Research Division:Information and Computing Sciences
Research Group:Artificial Intelligence and Image Processing
Research Field:Neural, Evolutionary and Fuzzy Computation
Objective Division:Information and Communication Services
Objective Group:Computer Software and Services
Objective Field:Computer Software and Services not elsewhere classified
Author:Xu, S (Dr Shuxiang Xu)
ID Code:81742
Year Published:2012
Deposited By:Computing and Information Systems
Deposited On:2013-01-04
Last Modified:2017-11-18
Downloads:0

Repository Staff Only: item control page