TY - JOUR
T1 - Training RBF network to tolerate single node fault
AU - Ho, Kevin
AU - Leung, Chi-sing
AU - Sum, John
PY - 2011/2/15
Y1 - 2011/2/15
N2 - In this paper, an objective function for training a radial basis function (RBF) network to handle single node open fault is presented. Based on the definition of this objective function, we propose a training method in which the computational complexity is the same as that of the least mean squares (LMS) method. Simulation results indicate that our method could greatly improve the fault tolerance of RBF networks, as compared with the one trained by LMS method. Moreover, even if the tuning parameter is misspecified, the performance deviation is not significant. © 2011 Elsevier B.V.
AB - In this paper, an objective function for training a radial basis function (RBF) network to handle single node open fault is presented. Based on the definition of this objective function, we propose a training method in which the computational complexity is the same as that of the least mean squares (LMS) method. Simulation results indicate that our method could greatly improve the fault tolerance of RBF networks, as compared with the one trained by LMS method. Moreover, even if the tuning parameter is misspecified, the performance deviation is not significant. © 2011 Elsevier B.V.
KW - Fault tolerant neural networks
KW - RBF
KW - Single node fault
UR - http://www.scopus.com/inward/record.url?scp=79751535157&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-79751535157&origin=recordpage
U2 - 10.1016/j.neucom.2010.12.005
DO - 10.1016/j.neucom.2010.12.005
M3 - RGC 21 - Publication in refereed journal
SN - 0925-2312
VL - 74
SP - 1046
EP - 1052
JO - Neurocomputing
JF - Neurocomputing
IS - 6
ER -