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

Attractor networks for shape recognition.

Y Amit1, M Mascaro

  • 1Department of Statistics, University of Chicago, Chicago, IL 60637, USA.

Neural Computation
|June 2, 2001
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

Proteomic analysis of the effect of hemin in breast cancer.

Scientific reports·2023
Same author

Iron cycle disruption by heme oxygenase-1 activation leads to a reduced breast cancer cell survival.

Biochimica et biophysica acta. Molecular basis of disease·2022
Same author

Transcriptomic response in thermally challenged seahorses Hippocampus erectus: The effect of magnitude and rate of temperature change.

Comparative biochemistry and physiology. Part B, Biochemistry & molecular biology·2022
Same author

Distribution patterns, carbon sources and niche partitioning in cave shrimps (Atyidae: Typhlatya).

Scientific reports·2020
Same author

Effect of a gradually increasing temperature on the behavioural and physiological response of juvenile Hippocampus erectus: Thermal preference, tolerance, energy balance and growth.

Journal of thermal biology·2019
Same author

Safety and efficacy of turoctocog alfa (NovoEight®) during surgery in patients with haemophilia A: results from the multinational guardian™ clinical trials.

Haemophilia : the official journal of the World Federation of Hemophilia·2014
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 a novel binary perceptron system capable of recognizing numerous shapes. It utilizes a recurrent network with Hebbian learning for stable, efficient shape recognition, offering biological parallels.

Area of Science:

  • Computational Neuroscience
  • Machine Learning
  • Pattern Recognition

Background:

  • Traditional neural networks often lack transparency and direct biological analogy.
  • Efficient shape recognition across a large number of classes remains a challenge in artificial intelligence.

Purpose of the Study:

  • To present a novel system of binary perceptrons for robust shape recognition.
  • To explore a recurrent network architecture with biologically plausible learning mechanisms.

Main Methods:

  • Utilized thousands of binary perceptrons with randomized feedforward connections and recurrent connections.
  • Implemented a Hebbian learning rule for modifying synaptic connections based on attractor activation and visual stimuli.
  • Trained the system by first activating class-specific attractors and then presenting visual input.

Related Experiment Videos

Main Results:

  • The system demonstrated the ability to recognize shapes across hundreds of classes.
  • Field-dependent Hebbian learning with positive synapses proved stable against variations in feature statistics and coding levels.
  • The network dynamics converged to sustained attractor states, enabling a form of working memory for recognition.

Conclusions:

  • The proposed perceptron system offers a transparent and biologically analogous alternative to standard feedforward networks.
  • The architecture supports robust, multi-class shape recognition through emergent network dynamics and stable learning.
  • This approach provides a foundation for developing more sophisticated biologically inspired computational models.