Title :
Toward the use of set-membership identification in efficient training of feedforward neural networks
Author :
Deller, J.R., Jr.
Author_Institution :
Dept. of Electr. Eng., Michigan State Univ., East Lansing, MI, USA
Abstract :
The application of the theory of set-membership identification to the development of efficient learning algorithms for neural networks is discussed. Some results relevant to the application of the method to nonlinear feedforward networks are presented. The techniques discussed have the potential to significantly improve the efficiency of teaching neural networks by employing novel data selection criteria based on set-theoretic constraints
Keywords :
learning systems; neural nets; set theory; data selection criteria; feedforward neural networks; learning algorithms; nonlinear feedforward networks; set-membership identification; set-theoretic constraints; teaching; Control systems; Feedforward neural networks; Intelligent networks; Neural networks; Samarium; Signal processing; Signal processing algorithms; Speech processing; Training data; Vectors;
Conference_Titel :
Circuits and Systems, 1990., IEEE International Symposium on
Conference_Location :
New Orleans, LA
DOI :
10.1109/ISCAS.1990.111974
