Self-Adaptive Particle Swarm Optimization-Based Echo State Network for Time Series Prediction

Author(s):  
Yu Xue ◽  
Qi Zhang ◽  
Ferrante Neri

Echo state networks (ESNs), belonging to the family of recurrent neural networks (RNNs), are suitable for addressing complex nonlinear tasks due to their rich dynamic characteristics and easy implementation. The reservoir of the ESN is composed of a large number of sparsely connected neurons with randomly generated weight matrices. How to set the structural parameters of the ESN becomes a difficult problem in practical applications. Traditionally, the design of the parameters of the ESN structure is performed manually. The manual adjustment of the ESN parameters is not convenient since it is an extremely challenging and time-consuming task. This paper proposes an ensemble of five particle swarm optimization (PSO) strategies to design the structure of ESN and then reduce the manual intervention in the design process. An adaptive selection mechanism is used for each particle in the evolution to select a strategy from the strategy candidate pool for evolution. In addition, leaky integration neurons are used as reservoir internal neurons, which are added within the adaptive mechanism for optimization. The root mean squared error (RMSE) is adopted as the evaluation criterion. The experimental results on Mackey–Glass time series benchmark dataset show that the proposed method outperforms other traditional evolutionary methods. Furthermore, experimental results on electrocardiogram dataset show that the proposed method on the ensemble of PSO displays an excellent performance on real-world problems.

2014 ◽  
Vol 511-512 ◽  
pp. 941-944 ◽  
Author(s):  
Hong Li Bian

Based on the particle swarm optimization (PSO) and BP neural network (BPNN), an algorithm for BP neural network optimized particle swarm optimization (PSOBPNN) is proposed. In the algorithm, PSO is used to obtain better network initial threshold and weight to compensate the defect of connection weight and thresholds of BPNN, thus it can make BPNN have faster convergence and greater learning ability. The efficiency of the proposed prediction method is tested by the simulation of the chaotic time series for Kent mapping. The simulations results show that the proposed method has higher forecasting accuracy compared with the BPNN, so it is proved that the algorithm is feasible and effective in the chaotic time series prediction.


2018 ◽  
Vol 51 (21) ◽  
pp. 219-223
Author(s):  
Lingshuang Kong ◽  
Xiaolong Gong ◽  
Chuanlai Yuan ◽  
Huiqin Xiao ◽  
Jianhua Liu

Sign in / Sign up

Export Citation Format

Share Document