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

Mixed states on neural network with structural learning.

Tomoyuki Kimoto1, Masato Okada

  • 1Oita National College of Technology, 1666 Maki, Oita-shi 870-0152, Japan. kimoto@oita-ct.ac.jp

Neural Networks : the Official Journal of the International Neural Network Society
|December 24, 2003
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

Comparison of the PK, PD, safety, tolerability, and immunogenicity of proposed biosimilar RGB-19 and tocilizumab in healthy Japanese males: a phase 1, randomised, crossover study.

EULAR rheumatology open·2026
Same author

Efficacy, safety, and immunogenicity of proposed biosimilar RGB-19 and tocilizumab intravenously administered to adults with active rheumatoid arthritis and an inadequate response to methotrexate: a phase 3, randomised study.

EULAR rheumatology open·2026
Same author

Impact of Diabetes on the Efficacy of Low-Voltage Area Ablation for Persistent Atrial Fibrillation: A Subanalysis of the SUPPRESS-AF Trial.

Journal of the American Heart Association·2026
Same author

Efficacy of Low-Voltage Area Ablation Across Substrate Size in Persistent Atrial Fibrillation: A Post Hoc Analysis of the SUPPRESS-AF Randomized Trial.

Journal of the American Heart Association·2026
Same author

Induction indices during ventricular pacing as an alternative diagnostic tool in supraventricular tachycardias.

Heart rhythm O2·2026
Same author

Triglyceride-Glucose Index and Clinical Outcome in Patients With Heart Failure With Preserved Ejection Fraction.

JACC. Asia·2026
Same journal

Dynamic analysis and reliable mechanical optimization application of ring HNN effected with a memristive neuron.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

DAFF-Net: A detection and search method for small-scale low surface brightness galaxies.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Quasi-synchronization for complex networks with hybrid pinning intermittent control.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Physics-encoded convolutional neural operators for parametric PDEs: A convergence-guaranteed framework via pre-computed kernel fields.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Exploiting audio-visual modalities in videos: Object detection via multi-stage bilateral coupling network.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Reliability-aware modality completion with cross-modal distillation for federated learning with missing modalities.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

This study explores mixed states in associative memory models, finding that memory patterns and mixed states easily coexist in sparse conditions. This research aids in developing transform-invariant recognition models.

Area of Science:

  • Computational neuroscience
  • Statistical mechanics
  • Machine learning

Background:

  • Associative memory models are crucial for understanding information storage and retrieval in neural networks.
  • Structural learning methods offer a way to dynamically form connections within these memory models.
  • Mixed states represent complex equilibria within memory systems, impacting their stability and capacity.

Purpose of the Study:

  • To investigate the properties of mixed states in a sparsely encoded associative memory model.
  • To analyze the coexistence and recall thresholds of memory patterns and mixed states.
  • To explore the implications of these findings for constructing transform-invariant recognition models.

Main Methods:

  • Utilized a structural learning method for associative memory model construction.

Related Experiment Videos

  • Employed statistical mechanical analysis to examine the properties of mixed states.
  • Investigated the behavior of the model in the sparse limit.

    • Main Results:

    • Identified 's' types of mixed states generated from 's' memory patterns as equilibrium states.
    • Demonstrated that storage capacity for memory patterns and specific mixed states diverge in the sparse limit.
    • Found that recall thresholds for memory patterns and mixed states are nearly equal, facilitating coexistence.

    Conclusions:

    • Memory patterns and mixed states can readily coexist in sparsely encoded associative memory models.
    • The observed properties are beneficial for developing transform-invariant recognition systems.
    • This work provides insights into the dynamics and capacity of complex memory networks.