Learning Convergence of CMAC Algorithm

Authors:
Chao He;Lixin Xu;Yuhe Zhang
Affiliations:
Department of Automatic Control, Beijing Institute of Technology, Beijing 100081, China. E-mail: chaohe@china.com;Department of Automatic Control, Beijing Institute of Technology, Beijing 100081, China. E-mail: chaohe@china.com;Department of Automatic Control, Beijing Institute of Technology, Beijing 100081, China. E-mail: chaohe@china.com
Venue:
Neural Processing Letters
Year:
2001

Citing 1
Cited 5

Advances in neural information processing systems 1

Advances in neural information processing systems 1

A General Learning Scheme for CMAC-based Controller

Neural Processing Letters
A Learning Algorithm of CMAC Based on RLS

Neural Processing Letters
Continuous CMAC-QRLS and Its Systolic Array

Neural Processing Letters
CMAC Neural Network and Its Use in Problems of Identification and Control of Nonlinear Dynamic Objects

Cybernetics and Systems Analysis
Gait Pattern Based on CMAC Neural Network for Robotic Applications

Neural Processing Letters

Quantified Score

Hi-index	0.00

Visualization

Abstract

CMAC convergence properties both in batch and in incremental learning are analyzed. The previous conclusions about the CMAC convergence, which are deduced under the condition that the articulation matrix is positive definite, are improved into the new less limited and more general conclusions in which no additive conditions are needed. An improved CMAC algorithm with self-optimizing learning rate is proposed from the new conclusions. Simulation results show the correctness of the new conclusions and the advantages of the improved algorithm.