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 Video

Updated: Jan 3, 2026

Visualizing Visual Adaptation
04:43

Visualizing Visual Adaptation

Published on: April 24, 2017

9.5K

An Adaptive Homeostatic Algorithm for the Unsupervised Learning of Visual Features.

Laurent U Perrinet1

  • 1INT, Inst Neurosci Timone, Aix Marseille Univ, CNRS, 27, Bd. Jean Moulin, CEDEX 5, 13385 Marseille, France.

Vision (Basel, Switzerland)
|November 19, 2019
PubMed
Summary

This study introduces a novel homeostasis-based regulation process for unsupervised learning in visual systems. This method enhances the efficient emergence of orientation-selective filters, applicable to machine learning and neural networks.

Related Concept Videos

Visual System01:26

Visual System

1.6K
Light enters the eye through the cornea, a transparent, dome-shaped surface covering the surface of the eyeball that helps to direct and focus incoming light. This light is then channeled toward the pupil, an adjustable opening whose size is controlled by the iris. The iris, a pigmented muscle, regulates the amount of light entering the eye by contracting or dilating the pupil, thereby ensuring optimal light levels for clear vision.
Once through the pupil, the light passes through the lens, a...
1.6K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Foveated Retinotopy Improves Classification and Localization in Convolutional Neural Networks.

Vision (Basel, Switzerland)·2026
Same author

A predictive approach to enhance time-series forecasting.

Nature communications·2025
Same author

Stakes of neuromorphic foveation: a promising future for embedded event cameras.

Biological cybernetics·2023
Same author

Beyond ℓ1 sparse coding in V1.

PLoS computational biology·2023
Same author

Learning heterogeneous delays in a layer of spiking neurons for fast motion detection.

Biological cybernetics·2023
Same author

Cortical recurrence supports resilience to sensory variance in the primary visual cortex.

Communications biology·2023

Area of Science:

  • Computational Neuroscience
  • Machine Learning
  • Systems Neuroscience

Background:

  • Visual system structure forms via unsupervised learning, creating feature-selective cells.
  • Hebbian learning models this, but struggles with simultaneous representation and encoder optimization.
  • Existing methods face challenges in balancing coding and learning steps.

Purpose of the Study:

  • To introduce a novel homeostasis-based regulation process for unsupervised learning.
  • To improve the efficiency of developing orientation-selective filters in neural networks.
  • To bridge biological insights with machine learning algorithms.

Main Methods:

  • Developed a homeostasis rule inspired by biological processes, using nonlinear adaptation.
  • Implemented a simplified heuristic based on neuron activation probability.
Keywords:
computer visionneurosciencesparsenessunsupervised learningvision

More Related Videos

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.9K

Related Experiment Videos

Last Updated: Jan 3, 2026

Visualizing Visual Adaptation
04:43

Visualizing Visual Adaptation

Published on: April 24, 2017

9.5K
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.9K
  • Utilized numerical simulations to compare optimal and heuristic homeostasis rules.

    • Main Results:

    • The proposed homeostasis rule facilitates efficient emergence of orientation-sensitive filters.
    • A simplified heuristic rule accelerates unsupervised learning while maintaining effectiveness.
    • Demonstrated application in convolutional neural networks for edge filter emergence.

    Conclusions:

    • Homeostasis offers a viable strategy for unsupervised learning in artificial systems.
    • The heuristic rule provides a faster, effective alternative to optimal homeostasis.
    • This approach enhances the development of feature detectors in machine learning models.