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

Linsker-type Hebbian Learning: A Qualitative Analysis on the Parameter Space.

Vwani P. Roychowdhury1, Hong Pan, Jianfeng Feng

  • 1School of Electrical and Computer Engineering, Purdue University, UK

Neural Networks : the Official Journal of the International Neural Network Society
|June 1, 1997
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

Magnetic phenolic resin core-shell structure derived carbon microspheres for ultrafast magnetic solid-phase extraction of triazine herbicides.

Journal of separation science·2022
Same author

fastDRH: a webserver to predict and analyze protein-ligand complexes based on molecular docking and MM/PB(GB)SA computation.

Briefings in bioinformatics·2022
Same author

High Expression of DC-STAMP Gene Predicts Adverse Outcomes in AML.

Frontiers in genetics·2022
Same author

Polyisocyanide Quaternary Ammonium Salts with Exceptionally Star-Shaped Structure for Enhanced Antibacterial Properties.

Polymers·2022
Same author

High recurrence of lymphedema and influencing factors in discharged breast cancer patients during the COVID-19 pandemic: A multicenter, cross-sectional survey.

Cancer medicine·2022
Same author

Ki67 for evaluating the prognosis of gastrointestinal stromal tumors: A systematic review and meta-analysis.

Oncology letters·2022
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

We developed a new method to analyze unsupervised learning in Linsker

Area of Science:

  • Computational Neuroscience
  • Machine Learning Theory

Background:

  • Linsker's multi-layer network model is a foundational model for understanding unsupervised learning.
  • The model's behavior is governed by nonlinear dynamics influenced by Hebbian learning and synaptic arbor density.
  • System parameters critically determine the emergence of specific receptive fields as stable attractors.

Purpose of the Study:

  • To establish a method for linking system parameters to unsupervised learning outcomes in Linsker's model.
  • To analyze the stability of connection patterns within the model's parameter space.
  • To elucidate the role of synaptic arbor density in receptive field stability.

Main Methods:

  • Derived a necessary and sufficient condition to assess the stability of saturated weight vectors.

Related Experiment Videos

  • Determined parameter regimes for the stability of specific connection patterns.
  • Utilized a parameter space approach to analyze receptive field stability.

    • Main Results:

    • Identified parameter ranges governing the stability of receptive fields in Linsker's network.
    • Demonstrated the significant impact of localized synaptic arbor density on pattern stability.
    • Provided a framework for analyzing stability in models using limiter functions for weight constraints.

    Conclusions:

    • The developed method offers a robust way to analyze unsupervised learning dynamics.
    • Synaptic arbor density is a critical factor in the stability of learned representations.
    • The approach is applicable to a broader class of learning and retrieval models.