The 'moving targets' training algorithm

Richard Rohwer, D. S. Touretzky (Editor)

    Research output: Contribution to conferencePaper

    Abstract

    A simple method for training the dynamical behavior of a neural network is derived. It is applicable to any training problem in discrete-time networks with arbitrary feedback. The algorithm resembles back-propagation in that an error function is minimized using a gradient-based method, but the optimization is carried out in the hidden part of state space either instead of, or in addition to weight space. Computational results are presented for some simple dynamical training problems, one of which requires response to a signal 100 time steps in the past.
    Original languageEnglish
    Publication statusUnpublished - 1990
    EventAdvances in Neural Information Processing Systems 1990 - Dublin, United Kingdom
    Duration: 29 Aug 199031 Aug 1990

    Other

    OtherAdvances in Neural Information Processing Systems 1990
    CountryUnited Kingdom
    CityDublin
    Period29/08/9031/08/90

    Keywords

    • dynamical behavior
    • neural network
    • error

    Fingerprint Dive into the research topics of 'The 'moving targets' training algorithm'. Together they form a unique fingerprint.

  • Cite this

    Rohwer, R., & Touretzky, D. S. (Ed.) (1990). The 'moving targets' training algorithm. Paper presented at Advances in Neural Information Processing Systems 1990, Dublin, United Kingdom.