Optimizing Deep Belief Echo State Network with a Sensitivity Analysis Input Scaling Auto-Encoder algorithm

作者：

Highlights：

• Optimizing the input weights and scaling parameters in DBESN by a novel SAIS-AE algorithm.

• Three widely used sequence tasks are applied to demonstrate the superiority of the proposed method.

• The proposed method significantly outperforms the DBESN and some other methods.

摘要

•Optimizing the input weights and scaling parameters in DBESN by a novel SAIS-AE algorithm.•Three widely used sequence tasks are applied to demonstrate the superiority of the proposed method.•The proposed method significantly outperforms the DBESN and some other methods.

论文关键词：Echo State Network,Reservoir computing,Recurrent neural network,Sensitivity Analysis Input Scaling,Auto-Encoder,Time-series prediction

论文评审过程：Received 10 March 2019, Revised 4 November 2019, Accepted 21 November 2019, Available online 25 November 2019, Version of Record 8 February 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2019.105257