Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Self-organized criticality and the self-organizing map.

J A Flanagan1

  • 1Neural Networks Research Center, Helsinki University of Technology, P.O. Box 5400, FIN-02015 HUT, Finland.

Physical Review. E, Statistical, Nonlinear, and Soft Matter Physics
|April 20, 2001
PubMed
Summary

The self-organizing map (SOM) demonstrates critical dynamics, functioning as a self-organized critical (SOC) model. This finding offers new perspectives on learning and topographic map formation in artificial neural networks.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Molecular and functional analysis identifies ALK-1 as the predominant cause of pulmonary hypertension related to hereditary haemorrhagic telangiectasia.

Journal of medical genetics·2003
Same author

Cloning of the cDNA for the putative calcium-sensing receptor and its tissue distribution in sea bream (Sparus aurata).

General and comparative endocrinology·2002
Same author

Calcium-sensing receptors and parathyroid hormone-related protein in the caudal neurosecretory system of the flounder (Platichthys flesus).

Journal of anatomy·2002
Same author

Self-organization in the one-dimensional SOM with a decreasing neighborhood.

Neural networks : the official journal of the International Neural Network Society·2002
Same author

Cloning of the cDNA for sea bream (Sparus aurata) parathyroid hormone-related protein.

General and comparative endocrinology·2000
Same author

Recombinant hyaluronate associated protein as a protective immunogen against Streptococcus equi and Streptococcus zooepidemicus challenge in mice.

Microbial pathogenesis·1999

Area of Science:

  • Artificial Neural Networks
  • Computational Neuroscience
  • Complex Systems

Background:

  • The self-organizing map (SOM) is a biologically inspired learning algorithm.
  • SOMs are known to converge to an ordered configuration, irrespective of initial conditions, particularly in one dimension.

Purpose of the Study:

  • To present the self-organizing map (SOM) as a self-organized critical (SOC) model.
  • To analyze and simulate the dynamics of the SOM in its ordered configuration.
  • To explore alternative interpretations of SOM learning and topographic map formation through the SOC lens.

Main Methods:

  • Theoretical analysis of SOM dynamics.
  • Computational simulation of SOM behavior.
  • Framing the SOM within the self-organized criticality (SOC) framework.

Related Experiment Videos

Main Results:

  • The dynamics of the SOM in its ordered configuration are shown to be critical.
  • The SOM can be effectively modeled as a SOC system.
  • Analysis and simulation confirm the critical nature of SOM dynamics.

Conclusions:

  • Viewing the SOM as a SOC system provides novel interpretations of its learning mechanisms.
  • The critical dynamics are fundamental to the formation of topographic maps.
  • This perspective enhances understanding of extremal dynamics in artificial neural networks.