HUSCAP logo Hokkaido Univ. logo

Hokkaido University Collection of Scholarly and Academic Papers >
Graduate School of Information Science and Technology / Faculty of Information Science and Technology >
Peer-reviewed Journal Articles, etc >

An Online Self-Constructive Normalized Gaussian Network with Localized Forgetting

Files in This Item:
e100-a_3_865.pdf1.29 MBPDFView/Open
Please use this identifier to cite or link to this item:http://hdl.handle.net/2115/65552

Title: An Online Self-Constructive Normalized Gaussian Network with Localized Forgetting
Authors: Backhus, Jana Browse this author
Takigawa, Ichigaku Browse this author →KAKEN DB
Imai, Hideyuki Browse this author →KAKEN DB
Kudo, Mineichi Browse this author →KAKEN DB
Sugimoto, Masanori Browse this author →KAKEN DB
Keywords: Normalized Gaussian Networks
dynamic model selection
online learning
chaotic time series forecasting
Issue Date: Mar-2017
Publisher: 電子情報通信学会
Journal Title: IEICE transactions on fundamentals of electronics communications and computer sciences
Volume: E100A
Issue: 3
Start Page: 865
End Page: 876
Publisher DOI: 10.1587/transfun.E100.A.865
Abstract: In this paper, we introduce a self-constructive Normalized Gaussian Network (NGnet) for online learning tasks. In online tasks, data samples are received sequentially, and domain knowledge is often limited. Then, we need to employ learning methods to the NGnet that possess robust performance and dynamically select an accurate model size. We revise a previously proposed localized forgetting approach for the NGnet and adapt some unit manipulation mechanisms to it for dynamic model selection. The mechanisms are improved for more robustness in negative interference prone environments, and a new merge manipulation is considered to deal with model redundancies. The effectiveness of the proposed method is compared with the previous localized forgetting approach and an established learning method for the NGnet. Several experiments are conducted for a function approximation and chaotic time series forecasting task. The proposed approach possesses robust and favorable performance in different learning situations over all testbeds.
Rights: copyright©2017 IEICE
Relation: http://search.ieice.org/
Type: article
URI: http://hdl.handle.net/2115/65552
Appears in Collections:情報科学院・情報科学研究院 (Graduate School of Information Science and Technology / Faculty of Information Science and Technology) > 雑誌発表論文等 (Peer-reviewed Journal Articles, etc)

Submitter: 杉本 雅則

Export metadata:

OAI-PMH ( junii2 , jpcoar_1.0 )

MathJax is now OFF:


 

 - Hokkaido University