The TAP approach to intensive and extensive connectivity systems

Yoshiyuki Kabashima, David Saad

    Research output: Chapter in Book/Published conference outputChapter

    Abstract

    The Thouless-Anderson-Palmer (TAP) approach was originally developed for analysing the Sherrington-Kirkpatrick model in the study of spin glass models and has been employed since then mainly in the context of extensively connected systems whereby each dynamical variable interacts weakly with the others. Recently, we extended this method for handling general intensively connected systems where each variable has only O(1) connections characterised by strong couplings. However, the new formulation looks quite different with respect to existing analyses and it is only natural to question whether it actually reproduces known results for systems of extensive connectivity. In this chapter, we apply our formulation of the TAP approach to an extensively connected system, the Hopfield associative memory model, showing that it produces identical results to those obtained by the conventional formulation.
    Original languageEnglish
    Title of host publicationAdvanced mean field methods: Theory and practice
    EditorsManfred Opper, David Saad
    Place of PublicationCambridge, US
    PublisherMIT
    Pages51-65
    Number of pages15
    ISBN (Print)0262150549
    Publication statusPublished - Feb 2001

    Publication series

    NameNeural Information Processing
    PublisherMassachusetts Institute of Technology Press (MIT Press)

    Bibliographical note

    Copyright of the Massachusetts Institute of Technology Press (MIT Press) Partially available on Google Books

    Keywords

    • Thouless-Anderson-Palmer
    • Sherrington-Kirkpatrick
    • general intensively connected systems
    • Hopfield associative memory model

    Fingerprint

    Dive into the research topics of 'The TAP approach to intensive and extensive connectivity systems'. Together they form a unique fingerprint.
    • Advanced mean field methods: theory and practice

      Opper, M. (Editor) & Saad, D. (Editor), Feb 2001, Cambridge, Massachusetts (US): MIT. 287 p. (Neural Information Processing)

      Research output: Book/ReportBook

    Cite this