Paper:
Views over last 60 days: 774
RasID - Random Search for Neural Network Training
Jinglu Hu, Kotaro Hirasawa and Junichi Murata
Department of Electrical and Electronic Systems Engineering, Kyushu University. 6-10-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan
Received:April 15, 1998Accepted:July 17, 1998Published:August 20, 1998
Keywords:Neural networks, Optimization, Random search, Multimodality, Intensified and diversified search
Abstract
This paper presents a novel random search, RasID, for neural network training, that introduces a sophisticated probability density function (PDF) in a random search for generating search vectors. The PDF provides two parameters to control local search ranges and directions efficiently. This realizes an intensified search where it is easy to find good solutions locally or a diversified search to escape local minima based on success-failure of past searches. Local gradients, if available, and trend information on the criterion function surface are used to improve search performance. The proposed scheme is applied to layered neural network training and is benchmarked against deterministic and nondeterministic methods.
Cite this article as:J. Hu, K. Hirasawa, and J. Murata, “RasID - Random Search for Neural Network Training,” J. Adv. Comput. Intell. Intell. Inform., Vol.2 No.4, pp. 134-141, 1998.Data files: