Title :
Discrete-time neuro identification without robust modification
Author_Institution :
Dept. de Control Autom., CINVESTAV-IPN, Mexico City, Mexico
fDate :
5/23/2003 12:00:00 AM
Abstract :
In general, neural networks cannot exactly represent nonlinear systems. A neuro-identifier has to include robust modification in order to guarantee Lyapunov stability. An input-to-state stability approach is used to create robust training algorithms for discrete-time neural networks. It is concluded that the gradient descent law and a backpropagation-type algorithm used for the weight adjustments are stable in the sense of L∞ and robust to any bounded uncertainties.
Keywords :
Lyapunov methods; backpropagation; discrete time systems; gradient methods; identification; neural nets; nonlinear systems; stability; Lyapunov stability; backpropagation-type algorithm; discrete-time system; gradient descent law; identification; input-to-state stability; neural networks; nonlinear system; robust training algorithms; weight adjustments;
Journal_Title :
Control Theory and Applications, IEE Proceedings -
DOI :
10.1049/ip-cta:20030204