Dynamic constructive fault tolerant algorithm for feedforward neural networks

被引：0

作者：

Hammadi, NC ^{[1
]}

Ohmameuda, T ^{[1
]}

Kaneko, K ^{[1
]}

Ito, H ^{[1
]}

机构：

[1] Chiba Univ, Grad Sch Sci & Technol, Chiba 263, Japan

来源：

IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS | 1998年 / E81D卷 / 01期

关键词：

feedforward neural network; dynamic constructive algorithm; fault tolerance; DCFTA;

D O I：

暂无

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this paper, a dynamic constructive algorithm for fault tolerant feedforward neural network, called DCFTA, is proposed. The algorithm starts with a network with single hidden neuron, and a new hidden unit is added dynamically to the network whenever it fails to converge. Before inserting the new hidden neuron into the network, only the weights connecting the new hidden neuron to the other neurons are trained (i.e., updated) until there is no significant reduction of the output error. To generate a fault tolerant network, the relevance of each synaptic weight is estimated in each cycle, and only the weights which have their relevance less than a specified threshold are updated in that cycle. The loss of a connections between neurons (which are equivalent to stuck-at-0 faults) are assumed. The simulation results indicate that the network constructed by DCFTA has a significant fault tolerance ability.

引用

页码：115 / 123

页数：9

共 50 条

[1] Fault tolerant constructive algorithm for feedforward neural networks
Hammadi, NC
Ohmameuda, T
Kaneko, K
Ito, H
[J]. PACIFIC RIM INTERNATIONAL SYMPOSIUM ON FAULT-TOLERANT SYSTEMS, PROCEEDINGS, 1997, : 215 - 220
[2] A learning algorithm for fault tolerant feedforward neural networks
Hammadi, C
Ito, H
[J]. IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 1997, E80D (01) : 21 - 27
[3] A constructive algorithm for feedforward neural networks
Institute of System Science, East China Normal University
不详
不详
[J]. 1600, 659-664 (2004):
[4] On fault tolerant training of feedforward neural networks
Arad, BS
ElAmawy, A
[J]. NEURAL NETWORKS, 1997, 10 (03) : 539 - 553
[5] A constructive algorithm for feedforward neural networks with incremental training
Liu, DR
Chang, TS
Zhang, Y
[J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS, 2002, 49 (12) : 1876 - 1879
[6] A constructive algorithm to synthesize arbitrarily connected feedforward neural networks
Puma-Villanueva, Wilfredo J.
dos Santos, Euripedes P.
Von Zuben, Fernando J.
[J]. NEUROCOMPUTING, 2012, 75 (01) : 14 - 32
[7] A CONSTRUCTIVE BASED HYBRID TRAINING ALGORITHM FOR FEEDFORWARD NEURAL NETWORKS
Ben Nasr, Mounir
Chtourou, Mohamed
[J]. 2009 6TH INTERNATIONAL MULTI-CONFERENCE ON SYSTEMS, SIGNALS AND DEVICES, VOLS 1 AND 2, 2009, : 97 - 100
[8] Constructive algorithm for fully connected cascade feedforward neural networks
Qiao, Junfei
Li, Fanjun
Han, Honggui
Li, Wenjing
[J]. NEUROCOMPUTING, 2016, 182 : 154 - 164
[9] THE TARGET SWITCH ALGORITHM - A CONSTRUCTIVE LEARNING PROCEDURE FOR FEEDFORWARD NEURAL NETWORKS
CAMPBELL, C
VICENTE, CP
[J]. NEURAL COMPUTATION, 1995, 7 (06) : 1245 - 1264
[10] Synthesis of fault-tolerant feedforward neural networks using minimax optimization
Deodhare, D
Vidyasagar, M
Keerthi, SS
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 1998, 9 (05): : 891 - 900

← 1 2 3 4 5 →