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

Density estimation by mixture models with smoothing priors

Utsugi1

  • 1National Institute of Bioscience and Human Technology, Ibaraki, JP, 1 1 Higashi Tsukuba, 305. utsugi@nibh.go.jp.

Neural Computation
|November 6, 1998
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

Observation of Multi-TeV Gamma Rays from the Crab Nebula using the Tibet Air Shower Array.

The Astrophysical journal·1999
Same journal

A Model-Free Reinforcement Learning Implementation of Decision Making Under Uncertainty by Sequential Sampling.

Neural computation·2026
Same journal

DROP: Distributional and Regular Optimism and Pessimism for Reinforcement Learning.

Neural computation·2026
Same journal

Hierarchical Active Inference Using Successor Representations.

Neural computation·2026
Same journal

W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators.

Neural computation·2026
Same journal

A Hidden Markov Model-Inspired Sequence Classification Method for Hyperdimensional Computing.

Neural computation·2026
Same journal

Sparse Graphical Modeling for Electrophysiological Phase-Based Connectivity Using Circular Statistics.

Neural computation·2026
See all related articles

This study introduces an extended self-organizing map (SOM) model with variable component selection probabilities. This enhanced model improves manifold coordinate system accuracy and density estimation compared to traditional SOMs.

Area of Science:

  • Machine Learning
  • Statistical Modeling
  • Data Visualization

Background:

  • Traditional self-organizing maps (SOMs) use fixed component selection probabilities, leading to centroid concentration in high-density areas.
  • This concentration deforms the coordinate system on extracted manifolds, hindering accurate smoothness evaluation.
  • Existing SOMs treat learning as a Gaussian mixture model estimation with a Gaussian smoothing prior on centroid parameters.

Purpose of the Study:

  • To address the limitations of fixed component selection probabilities in SOMs.
  • To develop an extended SOM model with variable component selection probabilities for improved manifold representation.
  • To enhance the accuracy of density estimation and manifold smoothness evaluation.

Main Methods:

Related Experiment Videos

  • Introduced an extended SOM model with variable component selection probabilities.
  • Incorporated a smoothing prior on component selection probabilities to stabilize estimation.
  • Developed an estimation algorithm for parameters and hyperparameters using empirical Bayesian inference.
  • Compared the performance of the new model and the traditional SOM model for density estimation via simulation experiments.

    • Main Results:

    • The extended SOM model demonstrates improved accuracy in representing manifold coordinate systems.
    • Variable component selection probabilities prevent centroid over-concentration in dense data regions.
    • Simulation experiments show the enhanced model's superior performance in density estimation compared to the standard SOM.
    • The introduced smoothing prior effectively stabilizes the estimation process.

    Conclusions:

    • The proposed extended SOM model with variable component selection probabilities offers a more accurate and robust approach to manifold learning and density estimation.
    • This model overcomes key limitations of traditional SOMs, particularly in handling varying data densities.
    • Empirical Bayesian inference provides a stable and effective method for parameter and hyperparameter estimation in the extended model.