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 supervised learning in networks with binary synapses.

Carlo Baldassi1, Alfredo Braunstein, Nicolas Brunel

  • 1Institute for Scientific Interchange Foundation, Viale S. Severo 65, I-10133 Torino, Italy.

Proceedings of the National Academy of Sciences of the United States of America
|June 22, 2007
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

Synaptic pruning, myelination and the emergence of psychiatric disorders in late adolescence.

bioRxiv : the preprint server for biology·2026
Same author

Correction: Inference in conditioned dynamics through causality restoration.

Scientific reports·2026
Same author

A Spatially Structured Spiking Network Model of Beta Traveling Waves and Their Attenuation in Motor Cortex.

bioRxiv : the preprint server for biology·2026
Same author

Climbing fibres recruit disinhibition to enhance Purkinje cell calcium signals.

Nature·2026
Same author

Corticothalamic communication for action coordination in a skilled motor behavior.

Nature neuroscience·2026
Same author

Sampling the space of solutions of an artificial neural network.

Physical review. E·2025
Same journal

Chemotactic self-organization captures the dynamics of mammalian hair follicle patterning.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Tomographic imaging of superconducting order using particle-hole interference.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Inhibitory potential of autologous neutralizing antibodies sets quantitative limits on the rebound-competent HIV-1 reservoir.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Inferring epidemiological parameters under an infectious phylogeography model with visitor dynamics.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Analytical modeling for suction cup designs for skin-interfaced wearable devices.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same journal

Improving cell-free metabolism through direct integration of artificial respiratory chains.

Proceedings of the National Academy of Sciences of the United States of America·2026
See all related articles

This study introduces an efficient online learning algorithm for discrete synaptic states, achieving near-theoretical learning limits in model neurons. The novel algorithm enhances robustness and offers potential for neurobiological and hardware implementation.

Area of Science:

  • Computational Neuroscience
  • Machine Learning
  • Synaptic Plasticity

Background:

  • Synaptic changes are discrete, posing computational challenges for learning.
  • Existing learning algorithms struggle with discrete synaptic states.

Purpose of the Study:

  • To develop and evaluate a neurobiologically plausible online learning algorithm for systems with discrete synapses.
  • To demonstrate the algorithm's efficiency and robustness in a model neuron learning task.

Main Methods:

  • Utilized a belief propagation-derived online learning algorithm.
  • Simulated a model neuron with binary synapses and finite hidden states learning a random classification task.
  • Analyzed synaptic transition rules and system robustness to noise.

Related Experiment Videos

Main Results:

  • The algorithm achieved near-theoretical learning limits, learning associations efficiently.
  • Optimal performance was observed with a finite number of hidden states, especially for sparse coding.
  • A metaplasticity rule stabilized synapses, and systems with more hidden states showed increased noise robustness.

Conclusions:

  • The developed algorithm is the first to efficiently achieve a finite number of learned patterns per binary synapse.
  • The algorithm's simplicity suggests potential for implementation in biological neural systems and hardware.
  • The findings advance understanding of learning mechanisms in discrete synaptic systems.