Stabilization of sequential learning neural network in sonar target classification via a novel approach

In this paper, the processing of sonar signals has been carried out using a Minimal Resource Allocation Network (MRAN) in identification of commonly encountered features in indoor environments. The stability-plasticity behaviors of the network have been investigated. From previous observations, the...

Full description

Bibliographic Details
Main Authors: Lim, , WS, Rao, , MVC
Format: Article
Published: 2005
Subjects:
Online Access:http://shdl.mmu.edu.my/2321/
_version_ 1848790025057075200
author Lim, , WS
Rao, , MVC
author_facet Lim, , WS
Rao, , MVC
author_sort Lim, , WS
building MMU Institutional Repository
collection Online Access
description In this paper, the processing of sonar signals has been carried out using a Minimal Resource Allocation Network (MRAN) in identification of commonly encountered features in indoor environments. The stability-plasticity behaviors of the network have been investigated. From previous observations, the experimental results show that MRAN possesses lower network complexity but experiences higher plasticity, and is unstable. A novel approach is proposed to solve these problems in MRAN and has also been experimentally proven that the network generalizes faster at a lower number of neurons (nodes) compared to the original MRAN. This new approach has been applied as a preprocessing tool to equip the network with certain information about the data to be used in training the network later. With this initial 'guidance', the network predicts extremely well in both sequential and random learning.
first_indexed 2025-11-14T18:06:03Z
format Article
id mmu-2321
institution Multimedia University
institution_category Local University
last_indexed 2025-11-14T18:06:03Z
publishDate 2005
recordtype eprints
repository_type Digital Repository
spelling mmu-23212011-08-24T02:53:50Z http://shdl.mmu.edu.my/2321/ Stabilization of sequential learning neural network in sonar target classification via a novel approach Lim, , WS Rao, , MVC QA75.5-76.95 Electronic computers. Computer science In this paper, the processing of sonar signals has been carried out using a Minimal Resource Allocation Network (MRAN) in identification of commonly encountered features in indoor environments. The stability-plasticity behaviors of the network have been investigated. From previous observations, the experimental results show that MRAN possesses lower network complexity but experiences higher plasticity, and is unstable. A novel approach is proposed to solve these problems in MRAN and has also been experimentally proven that the network generalizes faster at a lower number of neurons (nodes) compared to the original MRAN. This new approach has been applied as a preprocessing tool to equip the network with certain information about the data to be used in training the network later. With this initial 'guidance', the network predicts extremely well in both sequential and random learning. 2005 Article NonPeerReviewed Lim, , WS and Rao, , MVC (2005) Stabilization of sequential learning neural network in sonar target classification via a novel approach. NEURAL NETWORK WORLD, 15 I (2). pp. 111-127. ISSN 1210-0552
spellingShingle QA75.5-76.95 Electronic computers. Computer science
Lim, , WS
Rao, , MVC
Stabilization of sequential learning neural network in sonar target classification via a novel approach
title Stabilization of sequential learning neural network in sonar target classification via a novel approach
title_full Stabilization of sequential learning neural network in sonar target classification via a novel approach
title_fullStr Stabilization of sequential learning neural network in sonar target classification via a novel approach
title_full_unstemmed Stabilization of sequential learning neural network in sonar target classification via a novel approach
title_short Stabilization of sequential learning neural network in sonar target classification via a novel approach
title_sort stabilization of sequential learning neural network in sonar target classification via a novel approach
topic QA75.5-76.95 Electronic computers. Computer science
url http://shdl.mmu.edu.my/2321/