Growing RBF structures using self-organizing maps

Qingyu Xiong, Kotaro Hirasawa, Jinglu Hu, Junichi Murata

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

We present a novel growing RBF network structure using SOM in this paper. It consists of SOM and RBF networks respectively. The SOM performs unsupervised learning and also the weight vectors belonging to its output nodes are transmitted to the hidden nodes in the RBF networks as the centers of RBF activation functions, as a result one to one correspondence relationship is realized between the output nodes in SOM and the hidden nodes in RBF networks. The RBF networks perform supervised training using delta rule. Therefore, the current output errors in the RBF networks can be used to determine where to insert a new SOM unit according to the rule. This also makes it possible to make the RBF networks grow until a performance criterion is fulfilled or until a desired network size is obtained. The simulations on the two-spirals benchmark are shown to prove the proposed networks have good performance.

Original languageEnglish
Title of host publicationRobot and Human Communication - Proceedings of the IEEE International Workshop
Pages107-111
Number of pages5
Publication statusPublished - 2000
Event9th IEEE International Workshop on Robot and Human Interactive Communication RO-MAN2000 - Osaka, Japan
Duration: Sep 27 2000Sep 29 2000

Other

Other9th IEEE International Workshop on Robot and Human Interactive Communication RO-MAN2000
CountryJapan
CityOsaka
Period9/27/009/29/00

Fingerprint

Radial basis function networks
Self organizing maps
Unsupervised learning
Chemical activation

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Software

Cite this

Xiong, Q., Hirasawa, K., Hu, J., & Murata, J. (2000). Growing RBF structures using self-organizing maps. In Robot and Human Communication - Proceedings of the IEEE International Workshop (pp. 107-111)

Growing RBF structures using self-organizing maps. / Xiong, Qingyu; Hirasawa, Kotaro; Hu, Jinglu; Murata, Junichi.

Robot and Human Communication - Proceedings of the IEEE International Workshop. 2000. p. 107-111.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Xiong, Q, Hirasawa, K, Hu, J & Murata, J 2000, Growing RBF structures using self-organizing maps. in Robot and Human Communication - Proceedings of the IEEE International Workshop. pp. 107-111, 9th IEEE International Workshop on Robot and Human Interactive Communication RO-MAN2000, Osaka, Japan, 9/27/00.
Xiong Q, Hirasawa K, Hu J, Murata J. Growing RBF structures using self-organizing maps. In Robot and Human Communication - Proceedings of the IEEE International Workshop. 2000. p. 107-111
Xiong, Qingyu ; Hirasawa, Kotaro ; Hu, Jinglu ; Murata, Junichi. / Growing RBF structures using self-organizing maps. Robot and Human Communication - Proceedings of the IEEE International Workshop. 2000. pp. 107-111
@inproceedings{b42a50d67349408d8594c8e9997ca458,
title = "Growing RBF structures using self-organizing maps",
abstract = "We present a novel growing RBF network structure using SOM in this paper. It consists of SOM and RBF networks respectively. The SOM performs unsupervised learning and also the weight vectors belonging to its output nodes are transmitted to the hidden nodes in the RBF networks as the centers of RBF activation functions, as a result one to one correspondence relationship is realized between the output nodes in SOM and the hidden nodes in RBF networks. The RBF networks perform supervised training using delta rule. Therefore, the current output errors in the RBF networks can be used to determine where to insert a new SOM unit according to the rule. This also makes it possible to make the RBF networks grow until a performance criterion is fulfilled or until a desired network size is obtained. The simulations on the two-spirals benchmark are shown to prove the proposed networks have good performance.",
author = "Qingyu Xiong and Kotaro Hirasawa and Jinglu Hu and Junichi Murata",
year = "2000",
language = "English",
pages = "107--111",
booktitle = "Robot and Human Communication - Proceedings of the IEEE International Workshop",

}

TY - GEN

T1 - Growing RBF structures using self-organizing maps

AU - Xiong, Qingyu

AU - Hirasawa, Kotaro

AU - Hu, Jinglu

AU - Murata, Junichi

PY - 2000

Y1 - 2000

N2 - We present a novel growing RBF network structure using SOM in this paper. It consists of SOM and RBF networks respectively. The SOM performs unsupervised learning and also the weight vectors belonging to its output nodes are transmitted to the hidden nodes in the RBF networks as the centers of RBF activation functions, as a result one to one correspondence relationship is realized between the output nodes in SOM and the hidden nodes in RBF networks. The RBF networks perform supervised training using delta rule. Therefore, the current output errors in the RBF networks can be used to determine where to insert a new SOM unit according to the rule. This also makes it possible to make the RBF networks grow until a performance criterion is fulfilled or until a desired network size is obtained. The simulations on the two-spirals benchmark are shown to prove the proposed networks have good performance.

AB - We present a novel growing RBF network structure using SOM in this paper. It consists of SOM and RBF networks respectively. The SOM performs unsupervised learning and also the weight vectors belonging to its output nodes are transmitted to the hidden nodes in the RBF networks as the centers of RBF activation functions, as a result one to one correspondence relationship is realized between the output nodes in SOM and the hidden nodes in RBF networks. The RBF networks perform supervised training using delta rule. Therefore, the current output errors in the RBF networks can be used to determine where to insert a new SOM unit according to the rule. This also makes it possible to make the RBF networks grow until a performance criterion is fulfilled or until a desired network size is obtained. The simulations on the two-spirals benchmark are shown to prove the proposed networks have good performance.

UR - http://www.scopus.com/inward/record.url?scp=0034586686&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034586686&partnerID=8YFLogxK

M3 - Conference contribution

SP - 107

EP - 111

BT - Robot and Human Communication - Proceedings of the IEEE International Workshop

ER -