Research Article
Batch Gradient Training Method with Smoothing Regularization for Echo State Networks
107 downloads
@INPROCEEDINGS{10.1007/978-3-030-32388-2_42, author={Zohaib Ahmad and Kaizhe Nie and Junfei Qiao and Cuili Yang}, title={Batch Gradient Training Method with Smoothing Regularization for Echo State Networks}, proceedings={Machine Learning and Intelligent Communications. 4th International Conference, MLICOM 2019, Nanjing, China, August 24--25, 2019, Proceedings}, proceedings_a={MLICOM}, year={2019}, month={10}, keywords={Each state networks Gradient method regularization Sparsity}, doi={10.1007/978-3-030-32388-2_42} }- Zohaib Ahmad
Kaizhe Nie
Junfei Qiao
Cuili Yang
Year: 2019
Batch Gradient Training Method with Smoothing Regularization for Echo State Networks
MLICOM
Springer
DOI: 10.1007/978-3-030-32388-2_42
Abstract
The echo state networks (ESNs) have been widely used for time series prediction, due to their excellent learning performance and fast convergence speed. However, the obtained output weight of ESN by pseudoinverse is always ill-posed. In order to solve this problem, the ESN with batch gradient method and smoothing regularization (ESN-BGSL0) is studied. By introducing a smooth regularizer into the traditional error function, some redundant output weights of ESN-BGSL0 are driven to zeros and pruned. Two examples are performed to illustrate the efficiency of the proposed algorithm in terms of estimation accuracy and network compactness.
Copyright © 2019–2024 ICST