A characterization of simple recurrent neural networks with two hidden units as a language recognizer

Azusa Iwata, Yoshihisa Shinozawa, Akito Sakurai

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We give a necessary condition that a simple recurrent neural network with two sigmoidal hidden units to implement a recognizer of the formal language {a n b n | n > 0 } which is generated by a set of generating rules {S→aSb, S→ab } and show that by setting parameters so as to conform to the condition we get a recognizer of the language. The condition implies instability of learning process reported in previous studies. The condition also implies, contrary to its success in implementing the recognizer, difficulty of getting a recognizer of more complicated languages.

Original languageEnglish
Title of host publicationNeural Information Processing - 14th International Conference, ICONIP 2007, Revised Selected Papers
Pages436-445
Number of pages10
EditionPART 1
DOIs
Publication statusPublished - 2008
Event14th International Conference on Neural Information Processing, ICONIP 2007 - Kitakyushu, Japan
Duration: 2007 Nov 132007 Nov 16

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 1
Volume4984 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other14th International Conference on Neural Information Processing, ICONIP 2007
Country/TerritoryJapan
CityKitakyushu
Period07/11/1307/11/16

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'A characterization of simple recurrent neural networks with two hidden units as a language recognizer'. Together they form a unique fingerprint.

Cite this