Presentation 2005-01-28
Learning-Based Improvement in Fault Tolerance of Hopfield Associative Memories
Naotake KAMIURA, Teijiro ISOKAWA, Nobuyuki MATSUI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Hopfield neural networks tolerating weight faults are presented. The weight restriction and fault injection are adopted as fault-tolerant approaches. For the weight restriction, a range to which values of weights should belong is determined during the learning, and any weight being outside this range is forced to be either its upper limit or lower limit. A status of a fault occurring is then evoked by the fault injection, and calculating weights is made under this status. The learning based on both of the above approaches surpasses the learning based on either of them in the fault tolerance and/or in the learning time.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Hopfield Associative Memories / Fault Tolerance / Fault Injection / Weight Restriction / Hebbian Learning
Paper # CPM2004-172,ICD2004-217
Date of Issue

Conference Information
Committee ICD
Conference Date 2005/1/21(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 Integrated Circuits and Devices (ICD)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Learning-Based Improvement in Fault Tolerance of Hopfield Associative Memories
Sub Title (in English)
Keyword(1) Hopfield Associative Memories
Keyword(2) Fault Tolerance
Keyword(3) Fault Injection
Keyword(4) Weight Restriction
Keyword(5) Hebbian Learning
1st Author's Name Naotake KAMIURA
1st Author's Affiliation Graduate School of Engineering, University of Hyogo()
2nd Author's Name Teijiro ISOKAWA
2nd Author's Affiliation Graduate School of Engineering, University of Hyogo
3rd Author's Name Nobuyuki MATSUI
3rd Author's Affiliation Graduate School of Engineering, University of Hyogo
Date 2005-01-28
Paper # CPM2004-172,ICD2004-217
Volume (vol) vol.104
Number (no) 629
Page pp.pp.-
#Pages 6
Date of Issue