TY  - JOUR
Y1  - 2009///
JF  - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
A1  - Asirvadam, V.S.
UR  - https://www.scopus.com/inward/record.uri?eid=2-s2.0-76649131271&doi=10.1007%2f978-3-642-10677-4_14&partnerID=40&md5=4bb166c45741bee12c0b87061b0b951b
VL  - 5863 L
CY  - Bangkok
IS  - PART 1
N2  - A novel hybrid or separable recursive training strategies are de rived for the training of feedforward neural networks which incoporates a switching module. This new technique for updating weights combines non linear recursive training algorithms for the optimization of nonlinear weights with recursive least square type algorithms for the training of linear weights in one integrated routine. The proposed new variant of hybrid weight update includes switching mechanism based on the condition of input data to the system (correlated or noncorrelated). Simulation results demonstrate the im provement of the new proposed switching mode training scheme. © 2009 Springer-Verlag Berlin Heidelberg.
ID  - scholars610
KW  - FeedForward Network; Input datas; Non-linear; Perceptron; Recursive least squares; Recursive prediction; Recursive training algorithm; Simulation result; Switching mechanism; Switching modes; Switching modules; Training schemes; Training strategy; Weight update
KW  -  Data processing; Multilayers; Switching; Switching circuits; Switching systems
KW  -  Feedforward neural networks
SN  - 03029743
EP  - 134
AV  - none
N1  - cited By 0; Conference of 16th International Conference on Neural Information Processing, ICONIP 2009 ; Conference Date: 1 December 2009 Through 5 December 2009; Conference Code:79271
TI  - Separable recursive training algorithms with switching module
SP  - 126
ER  -