BMAC neural network and its application in function learning

Zhao Du Liu; Yue Feng Ma; Jing Bo Zhang; Zhi Quan Qi; Ming Song

BMAC neural network and its application in function learning

Zhao Du Liu^*, Yue Feng Ma, Jing Bo Zhang, Zhi Quan Qi, Ming Song

^*Corresponding author for this work

School of Mechanical Engineering

Beijing Institute of Technology

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

1 Citation (Scopus)

Abstract

BMAC neural network is established by the introduction of B-splines into CMAC neural network. The ways of establishing BMAC and the specialties of BMAC output and its receptive field functions are described in details. BMAC has continuous input and output space compared with CMAC whose input and output space are dispersed. In function learning, BMAC approaches faster and is more accurate than CMAC. At the same time, rules of influence of the studying parameters and weights initialization in function learning are obtained.

Original language	English
Pages (from-to)	858-861
Number of pages	4
Journal	Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology
Volume	24
Issue number	10
Publication status	Published - Oct 2004

Keywords

B-splines function
BMAC
Receptive field functions

Cite this

@article{77bebce404f4499cb0540fc0f99cbda7,

title = "BMAC neural network and its application in function learning",

abstract = "BMAC neural network is established by the introduction of B-splines into CMAC neural network. The ways of establishing BMAC and the specialties of BMAC output and its receptive field functions are described in details. BMAC has continuous input and output space compared with CMAC whose input and output space are dispersed. In function learning, BMAC approaches faster and is more accurate than CMAC. At the same time, rules of influence of the studying parameters and weights initialization in function learning are obtained.",

keywords = "B-splines function, BMAC, Receptive field functions",

author = "Liu, {Zhao Du} and Ma, {Yue Feng} and Zhang, {Jing Bo} and Qi, {Zhi Quan} and Ming Song",

year = "2004",

month = oct,

language = "English",

volume = "24",

pages = "858--861",

journal = "Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology",

issn = "1001-0645",

publisher = "Beijing Institute of Technology",

number = "10",

}

TY - JOUR

T1 - BMAC neural network and its application in function learning

AU - Liu, Zhao Du

AU - Ma, Yue Feng

AU - Zhang, Jing Bo

AU - Qi, Zhi Quan

AU - Song, Ming

PY - 2004/10

Y1 - 2004/10

N2 - BMAC neural network is established by the introduction of B-splines into CMAC neural network. The ways of establishing BMAC and the specialties of BMAC output and its receptive field functions are described in details. BMAC has continuous input and output space compared with CMAC whose input and output space are dispersed. In function learning, BMAC approaches faster and is more accurate than CMAC. At the same time, rules of influence of the studying parameters and weights initialization in function learning are obtained.

AB - BMAC neural network is established by the introduction of B-splines into CMAC neural network. The ways of establishing BMAC and the specialties of BMAC output and its receptive field functions are described in details. BMAC has continuous input and output space compared with CMAC whose input and output space are dispersed. In function learning, BMAC approaches faster and is more accurate than CMAC. At the same time, rules of influence of the studying parameters and weights initialization in function learning are obtained.

KW - B-splines function

KW - BMAC

KW - Receptive field functions

UR - http://www.scopus.com/inward/record.url?scp=11444257351&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:11444257351

SN - 1001-0645

VL - 24

SP - 858

EP - 861

JO - Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology

JF - Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology

IS - 10

ER -

BMAC neural network and its application in function learning

Abstract

Keywords

Other files and links

Fingerprint

Cite this