An adaptive Gauss-Newton algorithm for training multilayer nonlinear filters that have embedded memory (Q1971595)

From MaRDI portal
scientific article
Language Label Description Also known as
English
An adaptive Gauss-Newton algorithm for training multilayer nonlinear filters that have embedded memory
scientific article

    Statements

    An adaptive Gauss-Newton algorithm for training multilayer nonlinear filters that have embedded memory (English)
    0 references
    0 references
    7 June 2000
    0 references
    The authors consider multilayer dynamic Volterra networks. The networks consist of linear dynamic filters with nonlinear generalized single layer subnets. A modified Gauss-Newton optimization technique is used as a training algorithm. It is applied to modeling the inverse of a nonlinear dynamic tracking system.
    0 references
    neural networks
    0 references
    multilayer nonlinear adaptive filters
    0 references
    Gauss-Newton optimization training
    0 references
    inverse of a nonlinear dynamic tracking system
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references