DocumentCode
2927321
Title
Complexity reduction in Volterra connectionist modelling by consideration of output mapping
Author
Rayner, P. ; Lynch, M.
Author_Institution
Dept. of Eng., Cambridge Univ., UK
fYear
1990
fDate
3-6 Apr 1990
Firstpage
885
Abstract
The output mapping method is used to demonstrate an approach which makes the Volterra connectionist model highly efficient computationally, in comparison with current neural networks. Although the system was applied to yield linear recognizers, exactly the same approach can be used to reduce the order of a nonlinear recognizer. The network is not constrained to fit arbitrary indices and may utilize all of its degrees of freedom to improve recognition performance
Keywords
computational complexity; learning systems; neural nets; pattern recognition; Volterra connectionist modelling; linear recognizers; neural networks; nonlinear recognizer; output mapping; Adaptive filters; Computational modeling; Computer networks; Constraint optimization; Ear; Finite impulse response filter; Input variables; Intelligent networks; Kernel; Lagrangian functions; Neural networks; Pattern classification; Pattern recognition;
fLanguage
English
Publisher
ieee
Conference_Titel
Acoustics, Speech, and Signal Processing, 1990. ICASSP-90., 1990 International Conference on
Conference_Location
Albuquerque, NM
ISSN
1520-6149
Type
conf
DOI
10.1109/ICASSP.1990.115982
Filename
115982
Link To Document