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

Recursive self-organizing maps.

Thomas Voegtlin1

  • 1Institut des Sciences Cognitives, CNRS UMR 5015, Bron, France. voegtlin@isc.cnrs.fr

Neural Networks : the Official Journal of the International Neural Network Society
|November 6, 2002
PubMed
Summary
This summary is machine-generated.

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

Stability of fast oscillations in the mammalian olfactory bulb: experiments and modeling.

Journal of physiology, Paris·2011
Same author

Spike-timing error backpropagation in theta neuron networks.

Neural computation·2009
Same author

Adaptive synchronization of activities in a recurrent network.

Neural computation·2009
Same author

Recursive principal components analysis.

Neural networks : the official journal of the International Neural Network Society·2005
Same author

Linear recursive distributed representations.

Neural networks : the official journal of the International Neural Network Society·2005
Same author

Environmentally mediated synergy between perception and behaviour in mobile robots.

Nature·2003
Same journal

Dynamic analysis and reliable mechanical optimization application of ring HNN effected with a memristive neuron.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

DAFF-Net: A detection and search method for small-scale low surface brightness galaxies.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Quasi-synchronization for complex networks with hybrid pinning intermittent control.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Physics-encoded convolutional neural operators for parametric PDEs: A convergence-guaranteed framework via pre-computed kernel fields.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Exploiting audio-visual modalities in videos: Object detection via multi-stage bilateral coupling network.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Reliability-aware modality completion with cross-modal distillation for federated learning with missing modalities.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

This study introduces a novel nonlinear method to stabilize self-organizing maps (SOMs) with feedback, enabling them to learn sequential data. The enhanced SOM generalizes to represent input sequences recursively, adapting to temporal patterns.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Traditional neural networks use time-delayed feedback to implicitly represent temporal information.
  • Applying feedback to Self-Organizing Maps (SOMs) typically causes learning instability.
  • Representing sequential data is crucial for many AI and neuroscience applications.

Purpose of the Study:

  • To develop a stable feedback mechanism for Self-Organizing Maps (SOMs) to process sequential data.
  • To generalize SOMs for recursive representation of input sequences.
  • To adapt SOM representations to the temporal dynamics of input data.

Main Methods:

  • A novel nonlinearity was introduced to stabilize the learning process of SOMs with feedback.

Related Experiment Videos

  • The modified SOM was designed to learn and represent sequences recursively.
  • The temporal statistics of input series were analyzed to evaluate representation adaptation.

    • Main Results:

    • A stable, generalized SOM capable of learning from sequential inputs was successfully demonstrated.
    • The developed method overcomes the instability issue previously associated with SOM feedback.
    • The resulting SOM representations were shown to effectively adapt to the temporal structure of the input data.

    Conclusions:

    • The proposed nonlinear approach provides a stable and effective method for integrating feedback into SOMs for sequence representation.
    • This generalization of SOMs enhances their capability to model time-varying data.
    • The findings contribute to advancing unsupervised learning methods for temporal data analysis.