Title :
A training algorithm for SpikeProp improving stability of learning process
Author :
Toshiki, Wakamatsu ; Haruhiko, Takase ; Hiroharu, Kawanaka ; Shinji, Tsuruoka
Author_Institution :
Grad. Sch. of Eng., Mie Univ., Tsu, Japan
fDate :
July 31 2011-Aug. 5 2011
Abstract :
In this paper, we aims to improve stability of learning processes by the SpikeProp algorithm. We proposed the method that reduce the increase of the error in learning processes. It repeats two steps: (1) original SpikeProp algorithm, and (2) use a linear search in the steepest descent direction only if the first step is failed. Some experimental results shows the improvement of learning processes.
Keywords :
feedforward neural nets; learning (artificial intelligence); stability; SpikeProp algorithm; SpikeProp network; feed-forward networks; learning process; spiking neural networks; stability improvement; supervised learning algorithm; training algorithm; Biological neural networks; Iris; Iris recognition; Mathematical model; Neurons; Surges; Training;
Conference_Titel :
Neural Networks (IJCNN), The 2011 International Joint Conference on
Conference_Location :
San Jose, CA
Print_ISBN :
978-1-4244-9635-8
DOI :
10.1109/IJCNN.2011.6033325
