• LOGIN
Repository logo

BORIS Portal

Bern Open Repository and Information System

  • Publication
  • Projects
  • Funding
  • Research Data
  • Organizations
  • Researchers
  • LOGIN
Repository logo
Unibern.ch
  1. Home
  2. Publications
  3. Slow stochastic learning with global inhibition: a biological solution to the binary perceptron problem
 

Slow stochastic learning with global inhibition: a biological solution to the binary perceptron problem

Options
  • Details
BORIS DOI
10.48350/177109
Date of Publication
June 1, 2004
Publication Type
Article
Division/Institute

Institut für Physiolo...

Author
Senn, Walterorcid-logo
Institut für Physiologie
Fusi, Stefano
Institut für Physiologie
Subject(s)

600 - Technology::610...

Series
Neurocomputing
ISSN or ISBN (if monograph)
0925-2312
Publisher
Elsevier
Language
English
Publisher DOI
10.1016/j.neucom.2004.01.062
Description
Networks of neurons connected by plastic all-or-none synapses tend to quickly forget previously acquired information when new patterns are learned. This problem could be solved for random uncorrelated patterns by randomly selecting a small fraction of synapses to be modified upon each stimulus presentation (slow stochastic learning). Here we show that more complex, but still linearly separable patterns, can be learned by networks with binary excitatory synapses in a finite number of presentations provided that: (1) there is non-vanishing global inhibition, (2) the binary synapses are changed with small enough probability (slow learning) only when the output neuron does not give the desired response (as in the classical perceptron rule) and (3) the neuronal threshold separating the total synaptic inputs corresponding to different classes is small enough.
Related URL
https://www.sciencedirect.com/science/article/pii/S092523120400058X
Handle
https://boris-portal.unibe.ch/handle/20.500.12422/120297
Show full item
File(s)
FileFile TypeFormatSizeLicensePublisher/Copright statementContent
1-s2.0-S092523120400058X-main.pdftextAdobe PDF198.9 KBpublished
BORIS Portal
Bern Open Repository and Information System
Build: 360c85 [14.04. 8:05]
Explore
  • Projects
  • Funding
  • Publications
  • Research Data
  • Organizations
  • Researchers
More
  • About BORIS Portal
  • Send Feedback
  • Cookie settings
  • Service Policy
Follow us on
  • Mastodon
  • YouTube
  • LinkedIn
UniBe logo