The superior fault tolerance of artificial neural network training with a fault/noise injection- based genetic algorithm

Artificial neural networks （ANNs） are powerful compu- tational tools that are designed to replicate the human brain and adopted to solve a variety of problems in many different fields. Fault tolerance （FT）, an important property of ANNs, ensures their reliability when signifi- cant portions of a net...

Full description

Saved in:
Bibliographic Details
Published in:Protein & cell 2016-10, Vol.7 (10), p.735-748
Main Authors: Su, Feng, Yuan, Peijiang, Wang, Yangzhen, Zhang, Chen
Format: Article
Language:English
Subjects:
Algorithms
artificial neural networks
Biochemistry
Biomedical and Life Sciences
Cell Biology
Developmental Biology
fault tolerance
genetic algorithm
Human Genetics
Humans
Life Sciences
Models, Genetic
Neural Networks (Computer)
Protein Science
Research Article
Stem Cells
人工神经网络
噪声
容错神经网络
拟合误差
改进的BP算法
神经网络训练
节点故障
遗传算法
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Artificial neural networks （ANNs） are powerful compu- tational tools that are designed to replicate the human brain and adopted to solve a variety of problems in many different fields. Fault tolerance （FT）, an important property of ANNs, ensures their reliability when signifi- cant portions of a network are lost. In this paper, a fault/ noise injection-based （FIB） genetic algorithm （GA） is proposed to construct fault-tolerant ANNs. The FT per- formance of an FIB-GA was compared with that of a common genetic algorithm, the back-propagation algo- rithm, and the modification of weights algorithm. The FIB-GA showed a slower fitting speed when solving the exclusive OR （XOR） problem and the overlapping clas- sification problem, but it significantly reduced the errors in cases of single or multiple faults in ANN weights or nodes. Further analysis revealed that the fit weights showed no correlation with the fitting errors in the ANNs constructed with the FIB-GA, suggesting a relatively even distribution of the various fitting parameters. In contrast, the output weights in the training of ANNs implemented with the use the other three algorithms demonstrated a positive correlation with the errors. Our findings therefore Indicate that a combination of the fault/noise injection-based method and a GA is capable of introducing FT to ANNs and imply that the distributed ANNs demonstrate superior FT performance.
ISSN:1674-800X
1674-8018
DOI:10.1007/s13238-016-0302-5