Presentation 1994/11/11
Performance Analysis of SDAM with Feedback Circuit for a Hardware Neural Network
Tomochika Harada, Yoshihiro Hayakawa, Koji Nakajima, Yasuji Sawada,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The exploration of memorizing value of synaptic weight is necessary for a construction of hardware neural network.We have developped SDAM(Switched Diffusion Analog Memory)using non- volatile memory.The linearity of the change of synaptic weight is not perfect in SDAM.However,the linearity becomes perfect by adding a simple feedback circuit with SDAM.Pulse operation is superior as the method of neural network operation.SDAM is able to be operated in pulse made.The linearity of SDAM is independent of pulse duration.In this paper,We study the performance of SDAM with simple feedback circuit analytically and numerically.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) feedback
Paper # NLP94-67
Date of Issue

Conference Information
Committee NLP
Conference Date 1994/11/11(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 Nonlinear Problems (NLP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Performance Analysis of SDAM with Feedback Circuit for a Hardware Neural Network
Sub Title (in English)
Keyword(1) feedback
1st Author's Name Tomochika Harada
1st Author's Affiliation Research Institute of Electrical Communication Tohoku University()
2nd Author's Name Yoshihiro Hayakawa
2nd Author's Affiliation Research Institute of Electrical Communication Tohoku University
3rd Author's Name Koji Nakajima
3rd Author's Affiliation Research Institute of Electrical Communication Tohoku University
4th Author's Name Yasuji Sawada
4th Author's Affiliation Research Institute of Electrical Communication Tohoku University
Date 1994/11/11
Paper # NLP94-67
Volume (vol) vol.94
Number (no) 329
Page pp.pp.-
#Pages 5
Date of Issue