Izvestiya of Saratov University.

Mathematics. Mechanics. Informatics

ISSN 1816-9791 (Print)
ISSN 2541-9005 (Online)


For citation:

Slepovichev I. I. Algebraic properties of recurrent neural networks of discrete time. Izvestiya of Saratov University. Mathematics. Mechanics. Informatics, 2005, vol. 5, iss. 1, pp. 116-128. DOI: 10.18500/1816-9791-2005-5-1-116-128, EDN: IPKACW

This is an open access article distributed under the terms of Creative Commons Attribution 4.0 International License (CC-BY 4.0).
Published online: 
30.09.2005
Full text:
(downloads: 113)
Language: 
Russian
Heading: 
UDC: 
512.5
EDN: 
IPKACW

Algebraic properties of recurrent neural networks of discrete time

Autors: 
Slepovichev Ivan Ivanovich, Saratov State University
Abstract: 

Artificial neural networks can be used effectively for a quite general class of problems. Still there exists no formal foundation of some important constructions used in the theory. In this paper an attempt is undertaken to formalize some concepts of neuroinformatics and consider their properties from the point of view of applied universal algebra. It is proposed to treat neural networks as heterogeneous algebras which has made it possible to prove for them basic results similar to algebraic theorems on homomorphisms and congruences.

References: 
  1. McCulloc W. S., Pitts W. Н., “A logical calculus of the ideas immanent in nervous activity”, Bull. of Math. Biophysics, 5 (1943), 115–133
  2. Псиола В. В., “Обзор основных нейросетевых моделей”, Интеллектуальные системы, 4:3–4 (1999), 139–172
  3. Богомолов А. М., Салий В. Н., Алгебраические основы теории дискретных систем, М., 1997, 11
  4. Carrasco R. C., Mikel J. O., Forcada L., Efficient Encodings of finite automata in discrete-time recurrent neural networks
Received: 
17.03.2005
Accepted: 
16.08.2005
Published: 
30.09.2005