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

Efficient Biologically-based Pattern-recognizing Networks.

Moshe Gur1, Orly Yadid-Pecht

  • 1Technion-Israel Institute of Technology, Israel

Neural Networks : the Official Journal of the International Neural Network Society
|August 1, 1996
PubMed
Summary

This study introduces a biologically inspired neural network for pattern classification. It drastically reduces features, enabling compact, biologically plausible networks for efficient recognition, even with noisy data.

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

Seeing on the fly: No need for space-to-time encoding; saccade-generated transients enable fast, parallel representation of space.

Journal of vision·2025
Same author

Seeing on the fly: Physiological and behavioral evidence show that space-to-space representation and processing enable fast and efficient performance by the visual system.

Journal of vision·2024
Same author

There is a fundamental, unbridgeable gap between DNNs and the visual cortex.

The Behavioral and brain sciences·2023
Same author

Anatomical, physiological, and psychophysical data show that the nature of conscious perception is incompatible with the integrated information theory (IIT).

The Behavioral and brain sciences·2022
Same author

Very small faces are easily discriminated under long and short exposure times.

Journal of neurophysiology·2018
Same author

Wearable Microsystem for Minimally Invasive, Pseudo-Continuous Blood Glucose Monitoring: The e-Mosquito.

IEEE transactions on biomedical circuits and systems·2017

Area of Science:

  • Computational Neuroscience
  • Artificial Intelligence
  • Machine Learning

Background:

  • The human visual cortex employs sophisticated feature extraction for pattern recognition.
  • Existing machine-based recognition systems often lack biological plausibility and computational efficiency.

Purpose of the Study:

  • To develop a biologically motivated neural network for classification.
  • To create an automated feature selection process for improved recognition.
  • To enhance computational efficiency and biological plausibility in neural networks.

Main Methods:

  • A novel neural network architecture inspired by the visual cortex's feature extraction.
  • A feature ranking method based on discriminating ability to separate classes.
  • Automatic selection of optimal features for specific recognition tasks.

Related Experiment Videos

Main Results:

  • Drastic reduction in feature complexity.
  • Development of highly compact neural networks (tens of neurons).
  • Successful pattern classification even in noisy environments.

Conclusions:

  • The proposed method yields computationally efficient and biologically plausible neural networks.
  • Automated feature selection significantly reduces network complexity.
  • Compact neural networks can effectively classify patterns under challenging conditions.