Please use this identifier to cite or link to this item:
Title: Universal approximation and QoS violation application of extreme learning machine
Authors: Chen, L. 
Zhou, L. 
Pung, H.K. 
Keywords: Extreme learning machine (ELM)
Feedforward neural network
Radial basis function (RBF)
Randomhidden neurons
Universal approximation
Issue Date: 2008
Citation: Chen, L., Zhou, L., Pung, H.K. (2008). Universal approximation and QoS violation application of extreme learning machine. Neural Processing Letters 28 (2) : 81-95. ScholarBank@NUS Repository.
Abstract: Neural networks have been successfully applied to many applications due to their approximation capability. However, complicated network structures and algorithms will lead to computational and time-consuming burdens. In order to satisfy demanding real-time requirements, many fast learning algorithms were explored in the past. Recently, a fast algorithm, Extreme Learning Machine (ELM) (Huang et al. 70:489-501, 2006) was proposed. Unlike conventional algorithms whose neurons need to be tuned, the input-to-hidden neurons of ELM are randomly generated. Though a large number of experimental results have shown that input-to-hidden neurons need not be tuned, there lacks a rigorous proof whether ELM possesses the universal approximation capability. In this paper, based on the universal approximation property of an orthonormal method, we firstly illustrate the equivalent relationship between ELM and the orthonormal method, and further prove that neural networks with ELM are also universal approximations. We also successfully apply ELM to the identification of QoS violation in the multimedia transmission. © 2008 Springer Science+Business Media, LLC.
Source Title: Neural Processing Letters
ISSN: 13704621
DOI: 10.1007/s11063-008-9083-z
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.


checked on Sep 29, 2022


checked on Sep 29, 2022

Page view(s)

checked on Sep 22, 2022

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.