The Study of Fast Learning Method of Network

碩士 === 義守大學 === 電機工程學系碩士班 === 94 === The aim of this research is trying to solve and overcome the shortcomings of neural network with traditional error back-propagation (BP) learning algorithm. The modified learning algorithm is basically developed by using the technique of linear optimal solution a...

Full description

Bibliographic Details
Main Authors: Yu-Chuang Liu, 劉育昌
Other Authors: Rey-Chue Hwang
Format: Others
Language:zh-TW
Published: 2006
Online Access:http://ndltd.ncl.edu.tw/handle/92020499469415071408
id ndltd-TW-094ISU05442033
record_format oai_dc
spelling ndltd-TW-094ISU054420332015-10-13T14:49:54Z http://ndltd.ncl.edu.tw/handle/92020499469415071408 The Study of Fast Learning Method of Network 類神經網路快速學習之研究探討 Yu-Chuang Liu 劉育昌 碩士 義守大學 電機工程學系碩士班 94 The aim of this research is trying to solve and overcome the shortcomings of neural network with traditional error back-propagation (BP) learning algorithm. The modified learning algorithm is basically developed by using the technique of linear optimal solution and BP learning way. Such a combinational learning method not only make neural model have faster learning speed, but also make neural model avoid to plunging into local minimum. In addition, how to effectively use the linear optimal solution technique in finding the optimal weights of neural network is also studied and analyzed. Rey-Chue Hwang 黃瑞初 2006 學位論文 ; thesis 66 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 義守大學 === 電機工程學系碩士班 === 94 === The aim of this research is trying to solve and overcome the shortcomings of neural network with traditional error back-propagation (BP) learning algorithm. The modified learning algorithm is basically developed by using the technique of linear optimal solution and BP learning way. Such a combinational learning method not only make neural model have faster learning speed, but also make neural model avoid to plunging into local minimum. In addition, how to effectively use the linear optimal solution technique in finding the optimal weights of neural network is also studied and analyzed.
author2 Rey-Chue Hwang
author_facet Rey-Chue Hwang
Yu-Chuang Liu
劉育昌
author Yu-Chuang Liu
劉育昌
spellingShingle Yu-Chuang Liu
劉育昌
The Study of Fast Learning Method of Network
author_sort Yu-Chuang Liu
title The Study of Fast Learning Method of Network
title_short The Study of Fast Learning Method of Network
title_full The Study of Fast Learning Method of Network
title_fullStr The Study of Fast Learning Method of Network
title_full_unstemmed The Study of Fast Learning Method of Network
title_sort study of fast learning method of network
publishDate 2006
url http://ndltd.ncl.edu.tw/handle/92020499469415071408
work_keys_str_mv AT yuchuangliu thestudyoffastlearningmethodofnetwork
AT liúyùchāng thestudyoffastlearningmethodofnetwork
AT yuchuangliu lèishénjīngwǎnglùkuàisùxuéxízhīyánjiūtàntǎo
AT liúyùchāng lèishénjīngwǎnglùkuàisùxuéxízhīyánjiūtàntǎo
AT yuchuangliu studyoffastlearningmethodofnetwork
AT liúyùchāng studyoffastlearningmethodofnetwork
_version_ 1717758903715889152