Jove
Visualize
Contact Us

Related Experiment Videos

An evolutionary model of a neural network.

J C Parikh, R Pratap

    Journal of Theoretical Biology
    |May 7, 1984
    PubMed
    Summary
    This summary is machine-generated.

    This study proposes a general neural network model governed by an evolution equation. Specific assumptions allow this model to describe distributed memory with recognition and association, and even learning as a dynamic process.

    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

    Structure-based pharmacophore modelling for ErbB4-kinase inhibition: a systematic computational approach for small molecule drug discovery for breast cancer.

    SAR and QSAR in environmental research·2024
    Same author

    Liquid chromatography tandem mass spectrometry method for determination of antidiabetic chalcones derivative S001-469 in rat plasma, urine and feces: application to pharmacokinetic study.

    Drug research·2013
    Same author

    Bioanalytical LC-MS/MS method development and validation of novel antidiabetic candidate S007-1261 in rat plasma and its application to pharmacokinetic and oral bioavailability studies.

    Drug research·2013
    Same author

    Liquid chromatographic-tandem mass spectrometry assay for quantitation of a novel antidiabetic S002-853 in rat plasma and its application to pharmacokinetic study.

    Biomedical chromatography : BMC·2009
    Same author

    An unusual cause of obstructive sleep apnoea.

    The Journal of laryngology and otology·2009
    Same author

    Vagal nerve stimulator implantation: the otolaryngologist's perspective.

    European archives of oto-rhino-laryngology : official journal of the European Federation of Oto-Rhino-Laryngological Societies (EUFOS) : affiliated with the German Society for Oto-Rhino-Laryngology - Head and Neck Surgery·2008
    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

    Area of Science:

    • Computational Neuroscience
    • Artificial Intelligence
    • Dynamical Systems

    Background:

    • Neural networks are fundamental to understanding cognition and developing AI.
    • Existing models often lack a unified dynamical framework for memory and learning.
    • Distributed memory models are crucial for associative recall and pattern recognition.

    Purpose of the Study:

    • To propose a general, evolution-equation-based model for idealized neural networks.
    • To demonstrate how this model can represent distributed memory with recognition and association.
    • To show that the model can describe learning as a dynamic process.

    Main Methods:

    • Formulating a general evolution equation for neural network states.
    • Defining network evolution based on an initial state and a kernel function.

    Related Experiment Videos

  • Investigating specific kernel and initial state choices to derive memory and learning properties.

    • Main Results:

    • The evolution equation successfully models distributed memory with recognition and association under specific kernel assumptions.
    • This specific model aligns with existing Anderson and Cooper distributed memory models.
    • A different kernel choice extends the model to dynamically describe learning.

    Conclusions:

    • A unified dynamical framework for neural networks can capture both memory and learning.
    • The proposed evolution equation offers a flexible basis for future neural network research.
    • This work bridges idealized neural network theory with computational models of memory and learning.