Description and training of neural network dynamics

Richard Rohwer, F. Pasemann (Editor), H.D. Doebner (Editor)

    Research output: Unpublished contribution to conferenceUnpublished Conference Paperpeer-review

    Abstract

    Attractor properties of a popular discrete-time neural network model are illustrated through numerical simulations. The most complex dynamics is found to occur within particular ranges of parameters controlling the symmetry and magnitude of the weight matrix. A small network model is observed to produce fixed points, limit cycles, mode-locking, the Ruelle-Takens route to chaos, and the period-doubling route to chaos. Training algorithms for tuning this dynamical behaviour are discussed. Training can be an easy or difficult task, depending whether the problem requires the use of temporal information distributed over long time intervals. Such problems require training algorithms which can handle hidden nodes. The most prominent of these algorithms, back propagation through time, solves the temporal credit assignment problem in a way which can work only if the relevant information is distributed locally in time. The Moving Targets algorithm works for the more general case, but is computationally intensive, and prone to local minima.
    Original languageEnglish
    Publication statusPublished - 1991
    EventNeurodynamics, Proceedings of the 9th Summer Workshop - Arnold Sommerfeld Institute for Mathematical Physics, Clausthal, Germany
    Duration: 1 Jan 19911 Jan 1991

    Workshop

    WorkshopNeurodynamics, Proceedings of the 9th Summer Workshop
    Country/TerritoryGermany
    CityArnold Sommerfeld Institute for Mathematical Physics, Clausthal
    Period1/01/911/01/91

    Keywords

    • popular discrete-time neural
    • network model
    • simulations
    • weight matrix
    • algorithms
    • dynamical behaviour
    • temporal information
    • temporal credit assignment

    Fingerprint

    Dive into the research topics of 'Description and training of neural network dynamics'. Together they form a unique fingerprint.

    Cite this