Learning convergence of CMAC algorithm

Chao He; Lixin Xu; Yuhe Zhang

doi:10.1023/A:1011382225296

Learning convergence of CMAC algorithm

Chao He^*, Lixin Xu, Yuhe Zhang

^*Corresponding author for this work

School of Mechatronical Engineering

Beijing Institute of Technology

Research output: Contribution to journal › Article › peer-review

13 Citations (Scopus)

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.

Original language	English
Pages (from-to)	61-74
Number of pages	14
Journal	Neural Processing Letters
Volume	14
Issue number	1
DOIs	https://doi.org/10.1023/A:1011382225296
Publication status	Published - Aug 2001

Keywords

Batch learning
CMAC
Incremental learning
Learning convergence
Neural networks

Access to Document

10.1023/A:1011382225296

Cite this

@article{c757f4d5ab494ce1ae370067a7a971b5,

title = "Learning convergence of CMAC algorithm",

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.",

keywords = "Batch learning, CMAC, Incremental learning, Learning convergence, Neural networks",

author = "Chao He and Lixin Xu and Yuhe Zhang",

year = "2001",

month = aug,

doi = "10.1023/A:1011382225296",

language = "English",

volume = "14",

pages = "61--74",

journal = "Neural Processing Letters",

issn = "1370-4621",

publisher = "Springer Netherlands",

number = "1",

}

TY - JOUR

T1 - Learning convergence of CMAC algorithm

AU - He, Chao

AU - Xu, Lixin

AU - Zhang, Yuhe

PY - 2001/8

Y1 - 2001/8

N2 - 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.

AB - 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.

KW - Batch learning

KW - CMAC

KW - Incremental learning

KW - Learning convergence

KW - Neural networks

UR - http://www.scopus.com/inward/record.url?scp=0035426030&partnerID=8YFLogxK

U2 - 10.1023/A:1011382225296

DO - 10.1023/A:1011382225296

M3 - Article

AN - SCOPUS:0035426030

SN - 1370-4621

VL - 14

SP - 61

EP - 74

JO - Neural Processing Letters

JF - Neural Processing Letters

IS - 1

ER -

Learning convergence of CMAC algorithm

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this