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

Homeostatic synaptic scaling in self-organizing maps.

Thomas J Sullivan1, Virginia R de Sa

  • 1Department of Electrical Engineering, UC San Diego, 9500 Gilman Dr. MC 0515, 92093 La Jolla, CA, United States. tom@sullivan.to

Neural Networks : the Official Journal of the International Neural Network Society
|June 20, 2006
PubMed
Summary

Homeostatic synaptic scaling, a biological mechanism, can replace standard weight normalization in self-organizing maps (SOMs). This method supports map formation and neuron function during neural development, even with cell changes.

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

Deep learning for time-series segmentation of mechanical ventilator waveforms.

Scientific reports·2026
Same author

Finally, a use for balloons: Extended automated endovascular support enhances closed-loop resuscitation in a porcine model of shock.

The journal of trauma and acute care surgery·2026
Same author

Comparing Simultaneous Scalp EEG Recordings from the OpenBCI Cyton and Brain Products BrainAmp.

Sensors (Basel, Switzerland)·2026
Same author

Fostering Multidisciplinary Collaboration in Artificial Intelligence and Machine Learning Education: Tutorial Based on the AI-READI Bootcamp.

JMIR medical education·2025
Same author

Deep Learning for Time-Series Segmentation of Mechanical Ventilator Waveforms.

Research square·2025
Same author

Brain-Body Coupling in Listening to Metronomic Sounds and Music.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025

Area of Science:

  • Computational neuroscience
  • Artificial intelligence
  • Developmental neurobiology

Background:

  • Self-organizing maps (SOMs) are models of cortical development, typically using weight normalization.
  • Hebbian learning in these models often requires weight normalization to control growth.
  • Homeostatic plasticity, specifically synaptic scaling, is a biological mechanism observed in the cortex.

Purpose of the Study:

  • To investigate homeostatic synaptic scaling as an alternative to standard weight normalization in SOMs.
  • To determine if homeostatic synaptic scaling can support the formation of organized maps.
  • To assess if this mechanism allows neurons to maintain unsaturated firing rates.

Main Methods:

  • Simulated self-organizing maps (SOMs) using homeostatic synaptic scaling.

Related Experiment Videos

  • Compared the performance of SOMs with homeostatic synaptic scaling against those with standard weight normalization.
  • Analyzed neuron firing rates and map formation under conditions of cell proliferation or die-off.

    • Main Results:

    • Homeostatic synaptic scaling successfully replaces standard weight normalization in SOMs.
    • Organized maps still form, and output neurons maintain unsaturated firing rates.
    • Synaptic scaling can lead to networks that better represent input data probability distributions.

    Conclusions:

    • Homeostatic synaptic scaling is a viable and biologically plausible mechanism for controlling Hebbian learning in SOMs.
    • This mechanism supports robust map formation and neuronal function, adapting to changes in network size.
    • The findings suggest synaptic scaling may lead to more accurate data representation in neural network models.