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

Faithful representations with topographic maps.

M M. Van Hulle1

  • 1K.U. Leuven, Laboratorium voor Neuro-en Psychofysiologie, Faculteit Geneeskunde, Campus Gasthuisberg, Herestraat, B-3000, Leuven, Belgium

Neural Networks : the Official Journal of the International Neural Network Society
|March 29, 2003
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 journal

A boundary-regularization-enhanced video anomaly detection network based on context-adaptive spatio-temporal conditional diffusion.

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

MT<sup>2</sup>-CSD and LLM-CRAN: A new dataset and an LLM-based multi-semantic knowledge fusion model for conversational stance detection.

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

TriAlignNet: A triple-path cross-modality alignment framework for multimodal time series forecasting.

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

Anchor-based disentanglement framework for incremental multi-view clustering.

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

Complex-valued amplitude-phase interference modeling for adversarially robust classification.

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

TraNce: Type-aware hypergraph neural network with biological mediators for drug repositioning.

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

A new unsupervised learning rule, the kernel-based Maximum Entropy learning Rule (kMER), creates faithful topographic maps by directly optimizing information-theoretic entropy. This method improves upon existing techniques for applications like density estimation and adaptive filtering.

Area of Science:

  • Machine Learning
  • Computational Neuroscience
  • Information Theory

Background:

  • Topographic maps aim for faithful data representation, often striving for equiprobabilistic neuron activation.
  • Traditional methods like Mean Squared Error (MSE) minimization are not inherently suited for equiprobabilistic map formation.
  • Existing heuristics, such as adding a 'conscience' mechanism, attempt to bridge this gap in competitive learning.

Purpose of the Study:

  • Introduce a novel unsupervised competitive learning rule, the kernel-based Maximum Entropy learning Rule (kMER).
  • Develop a method that directly optimizes an information-theoretic criterion for topographic map formation.
  • Compare kMER's performance against existing rules for building equiprobabilistic maps.

Main Methods:

Related Experiment Videos

  • kMER associates a radially symmetric kernel with each neuron, updating both center and radius.
  • The learning rule maximizes the unconditional information-theoretic entropy of the neurons' outputs.
  • Benchmark tests involve two distribution types, comparing batch and incremental learning performance.

    • Main Results:

    • kMER demonstrates effective topographic map formation by directly optimizing information-theoretic entropy.
    • Performance benchmarks show kMER's capabilities in creating faithful representations.
    • The method is successfully applied to non-parametric density estimation and adaptive filtering of speech signals.

    Conclusions:

    • kMER offers a principled information-theoretic approach to unsupervised topographic map learning.
    • The developed topographic feature maps exhibit distinct characteristics compared to existing algorithms like Kohonen's SOM.
    • kMER shows promise for applications requiring accurate data representation and signal processing.