Learning convergence of CMAC algorithm

被引：12

作者：

He, C ^{[1
]}

Xu, LX ^{[1
]}

Zhang, YH ^{[1
]}

机构：

[1] Beijing Inst Technol, Dept Automat Control, Beijing 100081, Peoples R China

来源：

NEURAL PROCESSING LETTERS | 2001年 / 14卷 / 01期

关键词：

batch learning; CMAC; incremental learning; learning convergence; neural networks;

D O I：

10.1023/A:1011382225296

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 [模式识别与智能系统]; 0812 [计算机科学与技术]; 0835 [软件工程]; 1405 [智能科学与技术];

摘要：

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.

引用

页码：61 / 74

页数：14

共 16 条

[1]

Albus J. S., 1975, Transactions of the ASME. Series G, Journal of Dynamic Systems, Measurement and Control, V97, P220, DOI 10.1115/1.3426922

[2]

Albus J. S., 1975, Transactions of the ASME. Series G, Journal of Dynamic Systems, Measurement and Control, V97, P228, DOI 10.1115/1.3426923

[3]

CETINKUNT S, 1993, PROCEEDINGS OF THE 1993 AMERICAN CONTROL CONFERENCE, VOLS 1-3, P1976

[4]

DING L, 1997, NUMERICAL CALCULATIO

[5]

Fast tool servo control for ultra-precision machining at extremely low feed rates [J].

Ku, SS ;

Larsen, G ;

Cetinkunt, S .

MECHATRONICS, 1998, 8 (04) :381-393

[6]

CMAC neural network control for high precision motion control in the presence of large friction [J].

Larsen, GA ;

Cetinkunt, S ;

Donmez, A .

JOURNAL OF DYNAMIC SYSTEMS MEASUREMENT AND CONTROL-TRANSACTIONS OF THE ASME, 1995, 117 (03) :415-420

[7]

LARSEN GA, 1995, P ASME DYNAMIC SYSTE, V57, P497

[8]

LI B, 1999, P 1999 CHIN C ART IN, P823

[9]

LIU H, 1997, ACTA AUTOMATICA SINI, V23, P482

[10]

Luo Z.H., 1997, ACTA AUTOM SIN, V23, P455

← 1 2 →