Integration Method with Backpropagation

In this research, a new method is discovered (combined method) to accelerate the backpropagation network by using the expected values of source units for updating weights, we mean the expected value of unit by the sum of the output of the unit and its error term multiplied by the factor Beta to acce...

Full description

Saved in:
Bibliographic Details
Main Authors: Nidhal AL-Assady, Jamal Majeed, Shahbaa Khaleel
Format: Article
Language:English
Published: Mosul University 2005-06-01
Series:Al-Rafidain Journal of Computer Sciences and Mathematics
Subjects:
Online Access:https://csmj.mosuljournals.com/article_164073_b16b9f83822068ce0f9a43cf233c6846.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849727774169759744
author Nidhal AL-Assady
Jamal Majeed
Shahbaa Khaleel
author_facet Nidhal AL-Assady
Jamal Majeed
Shahbaa Khaleel
author_sort Nidhal AL-Assady
collection DOAJ
description In this research, a new method is discovered (combined method) to accelerate the backpropagation network by using the expected values of source units for updating weights, we mean the expected value of unit by the sum of the output of the unit and its error term multiplied by the factor Beta to accelerate the algorithm and also adjust the value of learning coefficient continuously if the value of energy function E decreases the learning rate is increased by a factor , if the value of the energy function E increases , the value of the learning rate is decreased by a factor . To obtain the optimal weight with minimum iteration and minimum time, we applied a new method on many applications to prove the result of this method (pattern compression, encoding and recognition on Arabic, English digits and alphabetic.
format Article
id doaj-art-bc4ba41c3a164d07a3e0732130f0a65b
institution DOAJ
issn 1815-4816
2311-7990
language English
publishDate 2005-06-01
publisher Mosul University
record_format Article
series Al-Rafidain Journal of Computer Sciences and Mathematics
spelling doaj-art-bc4ba41c3a164d07a3e0732130f0a65b2025-08-20T03:09:45ZengMosul UniversityAl-Rafidain Journal of Computer Sciences and Mathematics1815-48162311-79902005-06-0121496810.33899/csmj.2005.164073164073Integration Method with BackpropagationNidhal AL-Assady0Jamal Majeed1Shahbaa Khaleel2College of Computer Sciences and Mathematics University of Mosul, IraqCollege of Computer Sciences and Mathematics University of Mosul, IraqCollege of Computer Sciences and Mathematics University of MosulIn this research, a new method is discovered (combined method) to accelerate the backpropagation network by using the expected values of source units for updating weights, we mean the expected value of unit by the sum of the output of the unit and its error term multiplied by the factor Beta to accelerate the algorithm and also adjust the value of learning coefficient continuously if the value of energy function E decreases the learning rate is increased by a factor , if the value of the energy function E increases , the value of the learning rate is decreased by a factor . To obtain the optimal weight with minimum iteration and minimum time, we applied a new method on many applications to prove the result of this method (pattern compression, encoding and recognition on Arabic, English digits and alphabetic.https://csmj.mosuljournals.com/article_164073_b16b9f83822068ce0f9a43cf233c6846.pdfbackpropagationartificial neural networks
spellingShingle Nidhal AL-Assady
Jamal Majeed
Shahbaa Khaleel
Integration Method with Backpropagation
Al-Rafidain Journal of Computer Sciences and Mathematics
backpropagation
artificial neural networks
title Integration Method with Backpropagation
title_full Integration Method with Backpropagation
title_fullStr Integration Method with Backpropagation
title_full_unstemmed Integration Method with Backpropagation
title_short Integration Method with Backpropagation
title_sort integration method with backpropagation
topic backpropagation
artificial neural networks
url https://csmj.mosuljournals.com/article_164073_b16b9f83822068ce0f9a43cf233c6846.pdf
work_keys_str_mv AT nidhalalassady integrationmethodwithbackpropagation
AT jamalmajeed integrationmethodwithbackpropagation
AT shahbaakhaleel integrationmethodwithbackpropagation