Title :
Performance improvement of LMS algorithm using Hopfield model network
Author :
Takahasi, Kiyoshi ; Mori, Shinsaku
Author_Institution :
Dept. of Electr. Eng., Keio Univ., Yokohama, Japan
Abstract :
An algorithm that improves the adaptation rate of the least-mean-square algorithm and is based on the dynamics of the network in the Hopfield (neural network) model is discussed. The rate of adaptation of the algorithm is shown to be n times as fast as the system of the well-known LMS algorithm with the same control gain, n being the number of iterations for each data sample. The convergence is shown to depend on the gain constant, not on n. Simulations of the convergence behavior of the algorithm are presented
Keywords :
least squares approximations; neural nets; Hopfield model network; LMS algorithm; adaptation rate; control gain; convergence; data sample; gain constant; iterations; least-mean-square algorithm; network dynamics; neural network model; simulation; Adaptation model; Adaptive systems; Control systems; Convergence; Eigenvalues and eigenfunctions; Least squares approximation; Least squares methods; Neurons; Resonance light scattering; Vectors;
Conference_Titel :
Global Telecommunications Conference, 1990, and Exhibition. 'Communications: Connecting the Future', GLOBECOM '90., IEEE
Conference_Location :
San Diego, CA
Print_ISBN :
0-87942-632-2
DOI :
10.1109/GLOCOM.1990.116715
