Issue:Generalized net model of backpropagation learning algorithm

shortcut

http://ifigenia.org/wiki/issue:iwgn-2002-32-36

Title of paper: Generalized net model of backpropagation learning algorithm Author(s): Maciej Krawczak Systems Research Institute - Polish Academy of Sciences, Warsaw, Poland Wyzsza Szkola Informatyki Stosowanej i Zarzadzania krawczak@ibspan.waw.pl Hristo Aladjov CLBME - Bulg. Academy of Sciences, Acad. G. Bonchev Str., 105 Block, Sofia-1113, Bulgaria aladjov@clbme.bas.bg Presented at: 3rd IWGN, Sofia, 1 October 2002 Published in: Conference proceedings, pages 32—36 Download:  PDF (86  Kb, File info) Abstract: The backpropagation algorithm and its modifications are frequently used approaches for feed-forward neural networks training. It is shown in the present paper how generalized net model [1,2] of pure backpropagation algorithm can be constructed. The basic idea of this model is to be used as frame in which, with minor changes different algorithm definitions can be introduced, tested and compared. Keywords: Generalized nets, Neural networks, Machine learning, Backpropagation References: Atanassov K., Generalized Nets, World Scientific, Singapore, 1991. Atanassov K., Generalized index matrices, Comptes rendus de l'Academie Bulgare des Sciences, vol.40, 1987, No.11, 15-18. Atanassov, K., 1998. Generalized Nets in Artificial Intelligence. Vol. 1: Generalized nets and Expert Systems. "Prof. M. Drinov" Academic Publishing House, Sofia. Atanassov K., Aladjov, H., 2000. Generalized nets in artificial intelligence. Vol. 2, "Prof. Marin Drinov" Academic Publishing House, Sofia. Citations: The list of publications, citing this article may be empty or incomplete. If you can provide relevant data, please, write on the talk page.