Regresar

Sliding mode-based adaptive learning in dynamical-filter-weights neurons

Abstract:

A sliding mode control strategy is proposed for the synthesis of an adaptive learning algorithm in a neuron whose weights are constituted by first-order dynamical filters with adjustable parameters, which in turn allows the representation of dynamical processes in terms of a set of such neurons. The approach is shown to exhibit robustness characteristics and fast convergence properties. A simulation example, dealing with an analog signal tracking task, is provided which illustrates the feasibility of the approach.

Año de publicación:

2000

Keywords:

    Fuente:

    scopusscopus

    Tipo de documento:

    Article

    Estado:

    Acceso restringido

    Áreas de conocimiento:

    • Aprendizaje automático
    • Sistema de control

    Áreas temáticas:

    • Métodos informáticos especiales

    Contribuidores:

    Sira-Ramirez H.Sira-Ramirez H.
    Francklin RivasFrancklin Rivas
    Eliezer Colina-MorlesEliezer Colina-Morles