A Novel Learning Algorithm Which Makes Multilayer Neural Networks Multiple-Weight-Fault Tolerant

Itsuo TAKANAMI; Yasuhiro OYAMA

A Novel Learning Algorithm Which Makes Multilayer Neural Networks Multiple-Weight-Fault Tolerant

Itsuo TAKANAMI, Yasuhiro OYAMA

Full Text Views

0

Cite this

Summary :

We propose an efficient algorithm for making multi-layered neural networks (MLN) fault-tolerant to all multiple weight faults in a multi-dimensional interval by injecting intentionally two extreme multi-dimensional values in the interval into the weights of the selected multiple links in a learning phase. The degree of fault-tolerance to a multiple weight fault is measured by the number of essential multiple links. First, we analytically discuss how to choose effectively the multiple links to be injected, and present a learning algorithm for making MLNs fault tolerant to all multiple (i.e., simultaneous) faults in the interval defined by two multi-dimensional extreme points. Then it is proved that after the learning algorithm successfully finishes, MLNs become fault tolerant to all multiple faults in the interval. It is also shown that the time in a weight modification cycle depends little on multiplicity of faults k for small k. These are confirmed by simulation.

Publication: IEICE TRANSACTIONS on Information Vol.E86-D No.12 pp.2536-2543

Publication Date: 2003/12/01

Publicized

Online ISSN

DOI

Type of Manuscript: Special Section PAPER (Special Issue on Dependable Computing)

Category: Dependable Systems

Cite this

Copy

Itsuo TAKANAMI, Yasuhiro OYAMA, "A Novel Learning Algorithm Which Makes Multilayer Neural Networks Multiple-Weight-Fault Tolerant" in IEICE TRANSACTIONS on Information, vol. E86-D, no. 12, pp. 2536-2543, December 2003, doi: .
Abstract: We propose an efficient algorithm for making multi-layered neural networks (MLN) fault-tolerant to all multiple weight faults in a multi-dimensional interval by injecting intentionally two extreme multi-dimensional values in the interval into the weights of the selected multiple links in a learning phase. The degree of fault-tolerance to a multiple weight fault is measured by the number of essential multiple links. First, we analytically discuss how to choose effectively the multiple links to be injected, and present a learning algorithm for making MLNs fault tolerant to all multiple (i.e., simultaneous) faults in the interval defined by two multi-dimensional extreme points. Then it is proved that after the learning algorithm successfully finishes, MLNs become fault tolerant to all multiple faults in the interval. It is also shown that the time in a weight modification cycle depends little on multiplicity of faults k for small k. These are confirmed by simulation.
URL: https://global.ieice.org/en_transactions/information/10.1587/e86-d_12_2536/_p

Copy

@ARTICLE{e86-d_12_2536,
author={Itsuo TAKANAMI, Yasuhiro OYAMA, },
journal={IEICE TRANSACTIONS on Information},
title={A Novel Learning Algorithm Which Makes Multilayer Neural Networks Multiple-Weight-Fault Tolerant},
year={2003},
volume={E86-D},
number={12},
pages={2536-2543},
abstract={We propose an efficient algorithm for making multi-layered neural networks (MLN) fault-tolerant to all multiple weight faults in a multi-dimensional interval by injecting intentionally two extreme multi-dimensional values in the interval into the weights of the selected multiple links in a learning phase. The degree of fault-tolerance to a multiple weight fault is measured by the number of essential multiple links. First, we analytically discuss how to choose effectively the multiple links to be injected, and present a learning algorithm for making MLNs fault tolerant to all multiple (i.e., simultaneous) faults in the interval defined by two multi-dimensional extreme points. Then it is proved that after the learning algorithm successfully finishes, MLNs become fault tolerant to all multiple faults in the interval. It is also shown that the time in a weight modification cycle depends little on multiplicity of faults k for small k. These are confirmed by simulation.},
keywords={},
doi={},
ISSN={},
month={December},}

Copy

TY - JOUR
TI - A Novel Learning Algorithm Which Makes Multilayer Neural Networks Multiple-Weight-Fault Tolerant
T2 - IEICE TRANSACTIONS on Information
SP - 2536
EP - 2543
AU - Itsuo TAKANAMI
AU - Yasuhiro OYAMA
PY - 2003
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E86-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2003
AB - We propose an efficient algorithm for making multi-layered neural networks (MLN) fault-tolerant to all multiple weight faults in a multi-dimensional interval by injecting intentionally two extreme multi-dimensional values in the interval into the weights of the selected multiple links in a learning phase. The degree of fault-tolerance to a multiple weight fault is measured by the number of essential multiple links. First, we analytically discuss how to choose effectively the multiple links to be injected, and present a learning algorithm for making MLNs fault tolerant to all multiple (i.e., simultaneous) faults in the interval defined by two multi-dimensional extreme points. Then it is proved that after the learning algorithm successfully finishes, MLNs become fault tolerant to all multiple faults in the interval. It is also shown that the time in a weight modification cycle depends little on multiplicity of faults k for small k. These are confirmed by simulation.
ER -