DocumentCode :
2762289
Title :
Spiking neural network learning algorithms: Using learning rates adaptation of gradient and momentum steps
Author :
Delshad, Ehsan ; Moallem, P. ; Monadjemi, S A Hasan
Author_Institution :
Comput. Eng. Dept., Islamic Azad Univ., Arak, Iran
fYear :
2010
fDate :
4-6 Dec. 2010
Firstpage :
944
Lastpage :
949
Abstract :
In this paper we propose two learning algorithms for a spiking neural network which encodes information in the timing of spike trains. These algorithms are based on dynamic self adaptation for adapting the gradient learning rates (DS-η) and dynamic self adaptation for adapting the gradient learning rates and momentum (DS-ηα) algorithms. In our proposed algorithm, the optimum value for η was obtained from a parabolic function of error in both of these two algorithms and optimum value for α was obtained from our proposed adaptive algorithm. We performed a selection of benchmark problems to investigate the efficiency of our proposed algorithm. Compared to previously proposed algorithms such as SpikeProp and DS-ηα our algorithms, mod-DS-η and mod-DS-ηα, are faster than other methods in learning of the spiking neural networks.
Keywords :
gradient methods; learning (artificial intelligence); neural nets; dynamic self adaptation; gradient learning rates; spike trains; spiking neural network learning algorithms; Artificial neural networks; Computer architecture; Firing; Generators; Heuristic algorithms; Neurons; Optimized production technology; dynamic self adaptation; learning rate; local minimum; momentum; spiking neural network;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Telecommunications (IST), 2010 5th International Symposium on
Conference_Location :
Tehran
Print_ISBN :
978-1-4244-8183-5
Type :
conf
DOI :
10.1109/ISTEL.2010.5734158
Filename :
5734158
Link To Document :
بازگشت