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

Memory and Learning of Sequential Patterns by Nonmonotone Neural Networks.

Masahiko Morita1

  • 1University of Tsukuba, Japan

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

This study introduces a nonmonotone neural network (NNN) model for stable sequential pattern recall without synchronization. A novel learning algorithm embeds patterns in trajectory attractors, enabling smooth recall and efficient storage.

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

Topological modeling of gene expression in the brain with Huntington's disease reveals selective disruption of co-expression network.

Scientific reports·2026
Same author

Vascular effects of l-citrulline supplementation in healthy adults are largely influenced by the prehypertensive range of blood pressure: A double-blind, randomized, placebo-controlled, parallel-group study.

Nutrition research (New York, N.Y.)·2026
Same author

L-citrulline, Arginase Activity, and Nitrate As Potential Nitric Oxide-Related Markers of Subclinical Atherosclerosis in Healthy Middle-Aged Non-smokers: A Cross-Sectional Study.

Cureus·2025
Same author

Effects of Oral Ingestion of L-Ornithine on Mental Stress and Fatigue Based on the Trier Social Stress Test in Healthy Humans: A Randomized, Double-Blind, Placebo-Controlled, Parallel-Group Trial.

Journal of clinical medicine·2025
Same author

Spatial Statistics of Three-Dimensional Growth Dynamics of Spindle Microtubules.

Methods in molecular biology (Clifton, N.J.)·2024
Same author

Brainwave activities reflecting depressed mood: a pilot study.

Scientific reports·2023

Area of Science:

  • Computational Neuroscience
  • Artificial Intelligence
  • Dynamical Systems

Background:

  • Conventional neural networks struggle with temporal association tasks due to unsuitable dynamical properties for sequential pattern storage.
  • Lack of synchronizing neurons hinders the performance of existing models in recalling ordered information.

Purpose of the Study:

  • To introduce a novel nonmonotone neural network (NNN) model capable of storing and recalling sequential patterns.
  • To demonstrate stable and smooth recall of temporal sequences without relying on synchronizing neurons.

Main Methods:

  • Embedding sequential patterns within a trajectory attractor of the dynamical system.
  • Utilizing a simple covariance rule-based learning algorithm to modify synaptic weights.
  • Gradually varying input patterns to facilitate trajectory formation.

Related Experiment Videos

Main Results:

  • The NNN model achieves stable and smooth recall by having the network state follow the embedded trajectory attractor.
  • The learning algorithm effectively embeds sequential patterns with minimal repetitions.
  • The model demonstrates improved performance in temporal association tasks compared to conventional networks.

Conclusions:

  • The nonmonotone neural network offers a viable alternative for temporal association and sequential pattern processing.
  • The proposed learning algorithm is efficient and facilitates robust pattern embedding.
  • This model advances the understanding of neural network dynamics for sequential data handling.