Title :
Learning algorithms for cellular neural networks
Author :
Mirzai, Bahram ; Cheng, Zhenlan ; Moschytz, George S.
Author_Institution :
Signal & Inf. Process. Lab., Fed. Inst. of Technol., Zurich, Switzerland
Abstract :
A learning algorithm based on the decomposition of the A-template into symmetric and anti-symmetric parts is introduced. The performance of the algorithm is investigated in particular for coupled CNNs exhibiting diffusion-like and propagating behavior
Keywords :
cellular neural nets; learning (artificial intelligence); recurrent neural nets; A-template decomposition; anti-symmetric part; cellular neural networks; coupled CNNs; diffusion-like behavior; learning algorithms; propagating behavior; symmetric part; Backpropagation algorithms; Cellular neural networks; Cost function; Differential equations; Information processing; Laboratories; Minimization methods; Recurrent neural networks; Signal processing;
Conference_Titel :
Circuits and Systems, 1998. ISCAS '98. Proceedings of the 1998 IEEE International Symposium on
Conference_Location :
Monterey, CA
Print_ISBN :
0-7803-4455-3
DOI :
10.1109/ISCAS.1998.703935
