Presentation 2017-03-03
Hardware implementation of deep neural networks composed of self-organizing maps
Yuichiro Tanaka, Hakaru Tamukoh,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this research, we aim to implement deep neural networks (DNNs) composed of self-organizing maps into field programmable gate arrays (FPGAs) to develop a recognition system which is high speed and low power consumption. Reductions of multipliers from DNNs are required since multipliers in FPGA are limited and DNNs includes a lot of multipliers. We propose a hardware oriented algorithm for a DNN composed of self-organizing maps which reduces multipliers from DNNs. In addition, we develop a hardware of the DNN using the proposed hardware oriented algorithm and show simulation results.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Deep Neural Networks / Deep Learning / Self-Organizing Maps / FPGA
Paper # SIS2016-60
Date of Issue 2017-02-23 (SIS)

Conference Information
Committee SIS
Conference Date 2017/3/2(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Kanagawa Inst. Tech. Yokohama Office
Topics (in Japanese) (See Japanese page)
Topics (in English) Soft Computing, etc.
Chair Hirokazu Tanaka(Hiroshima City Univ.)
Vice Chair Takayuki Nakachi(NTT) / Noriaki Suetake(Yamaguchi Univ.)
Secretary Takayuki Nakachi(Toshiba) / Noriaki Suetake(Kanagawa Inst. of Tech.)
Assistant Hakaru Tamukoh(Kyushu Inst. of Tech.) / Masaaki Fujiyoshi(Tokyo Metropolitan Univ.)

Paper Information
Registration To Technical Committee on Smart Info-Media Systems
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Hardware implementation of deep neural networks composed of self-organizing maps
Sub Title (in English)
Keyword(1) Deep Neural Networks
Keyword(2) Deep Learning
Keyword(3) Self-Organizing Maps
Keyword(4) FPGA
1st Author's Name Yuichiro Tanaka
1st Author's Affiliation Kyushu Institute of Technology(KIT)
2nd Author's Name Hakaru Tamukoh
2nd Author's Affiliation Kyushu Institute of Technology(KIT)
Date 2017-03-03
Paper # SIS2016-60
Volume (vol) vol.116
Number (no) SIS-482
Page pp.pp.101-106(SIS),
#Pages 6
Date of Issue 2017-02-23 (SIS)