Presentation 2005/5/19
Hardware Neural Network for Real Time Learning : Hardware Neuro-system for Real Time Sensory External Check
Seungwoo CHUN, Yosihiro HAYAKAWA, Koji NAKAJIMA, Kouki HAGIWARA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) At factories, the external check of final products relies on the human check, which is based on experience and instinct. Such method is difficult to do long time and to reduce production cost. Therefore we expect an automatic external check system. We use the hardware neural network at the judgment enabling high-speed judgment and the automatic acquisition of the judgment criteria. In order to speed up operations, we have implemented parallel and pipeline processing into hardware. We design the circuit so that the derivative of the activation function is always a constant. The object time is within 0.1 second per one product. Measurements showing that the hardware neural network system can complete the process within 23ms are also presented.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Back-propagation / Sensory external check / PCI-BUS / Hardware
Paper # NC2005-6
Date of Issue

Conference Information
Committee NC
Conference Date 2005/5/19(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) Hardware Neural Network for Real Time Learning : Hardware Neuro-system for Real Time Sensory External Check
Sub Title (in English)
Keyword(1) Back-propagation
Keyword(2) Sensory external check
Keyword(3) PCI-BUS
Keyword(4) Hardware
1st Author's Name Seungwoo CHUN
1st Author's Affiliation Laboratory for Brainware/Laboratory for Nanoelectronics and Spintronics Research Institute of Electrical Communication, Tohoku University()
2nd Author's Name Yosihiro HAYAKAWA
2nd Author's Affiliation Laboratory for Brainware/Laboratory for Nanoelectronics and Spintronics Research Institute of Electrical Communication, Tohoku University
3rd Author's Name Koji NAKAJIMA
3rd Author's Affiliation Laboratory for Brainware/Laboratory for Nanoelectronics and Spintronics Research Institute of Electrical Communication, Tohoku University
4th Author's Name Kouki HAGIWARA
4th Author's Affiliation NEC Tohoku Manufacturing Systems. Co. Ltd.
Date 2005/5/19
Paper # NC2005-6
Volume (vol) vol.105
Number (no) 82
Page pp.pp.-
#Pages 6
Date of Issue