Presentation 2011-01-24
Investigate of Multi-Layer Perceptron with Glial Network Receiving Local External Stimulus
Chihiro IKUTA, Yoko UWATE, Yoshifumi NISHIO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We have proposed the glial network which was inspired from the feature of brain. In the glial network, glias generate independent oscillations and these oscillations propagated neurons and other glias. We confirmed that the glial network improved the learning performance of the Multi-Layer Perceptron (MLP) by connecting the neurons more effectively than the conventional networks. In this article, we investigate the glial network when the glias do not oscillate by themselves but only one glia is stimulated by an external noise. The external noise propagates the glial network and influences to the MLP. By the effect of the one-way influence via the glial network, a kind of position-depending-feature appears in the MLP. The simulation results show that the proposed networks possesses better learning performance, more biased anti-damaging property, and better generalization capability than the conventional networks.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Glial network / MLP / External stimulus
Paper # NLP2010-126,NC2010-90
Date of Issue

Conference Information
Committee NC
Conference Date 2011/1/17(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Investigate of Multi-Layer Perceptron with Glial Network Receiving Local External Stimulus
Sub Title (in English)
Keyword(1) Glial network
Keyword(2) MLP
Keyword(3) External stimulus
1st Author's Name Chihiro IKUTA
1st Author's Affiliation Dept. of E. E. Eng., Tokushima University()
2nd Author's Name Yoko UWATE
2nd Author's Affiliation Dept. of E. E. Eng., Tokushima University
3rd Author's Name Yoshifumi NISHIO
3rd Author's Affiliation Dept. of E. E. Eng., Tokushima University
Date 2011-01-24
Paper # NLP2010-126,NC2010-90
Volume (vol) vol.110
Number (no) 388
Page pp.pp.-
#Pages 5
Date of Issue