Title :
Continuation-based learning algorithm for discrete-time cellular neural networks
Author :
Magnussen, Holger ; Papoutsis, Georgiog ; Nossek, Josef A.
Author_Institution :
Inst. for Network Theory & Circuit Design, Tech. Univ. Munchen, Germany
Abstract :
The SGN-type nonlinearity of a standard discrete-time cellular neural network (DTCNN) is replaced by a smooth, sigmoidal nonlinearity with variable gain. Therefore, the resulting dynamical system is fully differentiable. Bounds on gain of the sigmoidal function are given, so that the new smooth system approximates the standard DTCNN within certain limits. A learning algorithm is proposed, which finds the template parameters for the standard DTCNN by gradually increasing the gain of the sigmoidal function
Keywords :
cellular neural nets; learning (artificial intelligence); continuation-based learning algorithm; discrete-time cellular neural networks; sigmoidal function; sigmoidal nonlinearity; template parameters; Cellular networks; Cellular neural networks; Circuit synthesis; Electronic mail; Integrated circuit interconnections;
Conference_Titel :
Cellular Neural Networks and their Applications, 1994. CNNA-94., Proceedings of the Third IEEE International Workshop on
Conference_Location :
Rome
Print_ISBN :
0-7803-2070-0
DOI :
10.1109/CNNA.1994.381689
