Title :
A geometric method to obtain error-correcting classification by neural networks with fewer hidden units
Author :
V. Kurkova;P.C. Kainen
Author_Institution :
Inst. of Comput. Sci., Acad. of Sci., Prague, Czech Republic
Abstract :
Coding the output of a neural network can reduce the output dimension but may also increase the requirements for approximation accuracy tolerated by error-correcting. Using a generalization of orthogonality, we propose a method for error-correcting classification by feedforward neural networks that reduces the number of hidden units for a variety of classes of patterns.
Keywords :
"Neural networks","Feedforward neural networks","Multi-layer neural network","Computer errors","Computer science","Electronic mail","Computer industry","Postal services","Mouth","Face recognition"
Conference_Titel :
Neural Networks, 1996., IEEE International Conference on
Print_ISBN :
0-7803-3210-5
DOI :
10.1109/ICNN.1996.549073
