论文信息 - Backpropagation learning with a (1+1) ES

Backpropagation learning with a (1+1) ES

We present an improvement to backpropagation (BP) learning for neural networks using a (1+1) evolutionary strategy (ES). The goal is to provide a method than can adaptively change the learning parameters used in BP in an unconstrained manner. The BP/ES algorithm we propose is simple to implement and can be used with various versions of BP. In our experiments there is a substantial increase in performance for time series prediction

Patricia Melin | Leonardo Trujillo | José Parra Galaviz

[1] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .

[2] Sandhya Samarasinghe,et al. Neural Networks for Applied Sciences and Engineering , 2006 .

[3] Robert A. Jacobs,et al. Increased rates of convergence through learning rate adaptation , 1987, Neural Networks.

[4] Patrick van der Smagt,et al. Introduction to neural networks , 1995, The Lancet.

[5] Juan Manuel Corchado Rodríguez. Redes neuronales artificiales: un enfoque práctico , 2000 .

[6] Martin T. Hagan,et al. Neural network design , 1995 .