Optimization of the weights and asymmetric activation function family of neural network for time series forecasting

作者：

Highlights：

• We present a method for optimization of the activation functions for ANN.

• The proposed optimization method uses Simulated Annealing and Tabu Search.

• The proposed method is good for forecasting distinct time series.

摘要

•We present a method for optimization of the activation functions for ANN.•The proposed optimization method uses Simulated Annealing and Tabu Search.•The proposed method is good for forecasting distinct time series.

论文关键词：Neural networks,Asymmetric activation function,Free parameter,Simulated annealing,Tabu search,BPM algorithm,LM algorithm,Time series

论文评审过程：Available online 30 May 2013.

论文官网地址：https://doi.org/10.1016/j.eswa.2013.05.053