Jove
Visualize
Contact Us

Related Experiment Videos

Neural maps and topographic vector quantization.

H -U. Bauer1, M Herrmann, T Villmann

  • 1MPI für Strömungsforschung, Bunsenstrasse 10, 37073, Göttingen, Germany

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

Quantifying topography in neural maps is challenging but crucial for applications like data visualization. New measures assess map topography, proving adequate for selecting optimal dimensions and identifying non-topographic maps.

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

Live-cell imaging reveals decreased cAMP in a PFE-associated c.1050-3C>G PTH1R cell model.

Journal of molecular medicine (Berlin, Germany)·2026
Same author

Ecosystem Metabolic Rates Estimated from Diel Oxygen Measurements in Two Subtropical Estuaries.

Estuaries and coasts : journal of the Estuarine Research Federation·2025
Same author

It is not waste if it is therapy: cellular, secretory and functional properties of reamer-irrigator-aspirator (RIA)-derived autologous bone grafts.

Journal of orthopaedics and traumatology : official journal of the Italian Society of Orthopaedics and Traumatology·2025
Same author

Prodromal or mild Alzheimer's disease: Influence of neuropsychiatric symptoms and premordid personality on caregivers' burden.

International journal of geriatric psychiatry·2024
Same author

Determinants of circulating calcitriol in cardiovascular disease.

The Journal of steroid biochemistry and molecular biology·2024
Same author

Review: cellularity in bone marrow autografts for bone and fracture healing.

American journal of physiology. Cell physiology·2023
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

Area of Science:

  • Computational neuroscience
  • Machine learning
  • Data analysis

Background:

  • Neural maps integrate data representation via codebook vectors with topographic properties, similar to continuous functions.
  • While quantization error is easily measured, quantifying map topography remains difficult.
  • Topography is advantageous for applications such as noise reduction in transmission channels and data visualization.

Purpose of the Study:

  • To review conceptual definitions of topography in neural maps.
  • To introduce and evaluate recently proposed measures for quantifying map topography.
  • To assess the utility of these measures on synthetic and real-world datasets.

Main Methods:

  • Conceptual review of topographic definitions and quantification methods.

Related Experiment Videos

  • Application of proposed topographic measures to neural maps trained on synthetic data.
  • Testing measures on neural maps generated from real-world datasets: chaotic time series, speech, and image data.

    • Main Results:

    • Measures were evaluated for reproducibility, scalability, and dependence on map topology.
    • The proposed measures adequately identified the topographically optimal output space dimension for real-world data.
    • The measures consistently distinguished between topographic and non-topographic neural maps.

    Conclusions:

    • Quantifying neural map topography is feasible with the proposed measures, despite inherent challenges.
    • These measures offer practical utility in selecting optimal map configurations and identifying suboptimal ones.
    • Further refinement of topographic measures can enhance their performance in diverse applications.