Echo state networks with double-reservoir for time-series prediction

Chong Liu; Huaguang Zhang; Xianshuang Yao; Kun Zhang

doi:10.1109/ICICIP.2016.7885901

Echo state networks with double-reservoir for time-series prediction

Chong Liu
, Huaguang Zhang
, Xianshuang Yao
, Kun Zhang

Northeastern University China

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

6 Scopus citations

Abstract

In this paper, a novel model, named double-reservoir echo state networks (DR-ESN), is proposed. DR-ESN is constructed by two reservoirs which are connected in series, thus the performance of abstracting the characteristics from the prediction task is improved. A sufficient condition is provided to ensure the stability of DR-ESN. The batch gradient method and ridge regression method are utilized to optimize the six parameters of DR-ESN and train the readouts, respectively. DR-ESN is verified by two different experiments, chaotic time series prediction and real-valued function time series prediction. The simulation results demonstrates that DR-ESN has a more precise result than leaky-ESN in predicting the time series.

Original language	English
Title of host publication	7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	196-202
Number of pages	7
ISBN (Electronic)	9781509021550
DOIs	https://doi.org/10.1109/ICICIP.2016.7885901
State	Published - 23 Mar 2017
Externally published	Yes
Event	7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Siem Reap, Cambodia Duration: 1 Dec 2016 → 4 Dec 2016

Publication series

Name	7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings

Conference

Conference	7th International Conference on Intelligent Control and Information Processing, ICICIP 2016
Country/Territory	Cambodia
City	Siem Reap
Period	1/12/16 → 4/12/16

Keywords

Batch gradient descent
Double-reservoir
Echo state networks
Time-series prediction

Access to Document

10.1109/ICICIP.2016.7885901

Cite this

Liu, C., Zhang, H., Yao, X., & Zhang, K. (2017). Echo state networks with double-reservoir for time-series prediction. In 7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings (pp. 196-202). Article 7885901 (7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICICIP.2016.7885901

Liu, Chong ; Zhang, Huaguang ; Yao, Xianshuang et al. / Echo state networks with double-reservoir for time-series prediction. 7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 196-202 (7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings).

@inproceedings{73b09b8976c744faae5f0dcacc02e2dc,

title = "Echo state networks with double-reservoir for time-series prediction",

abstract = "In this paper, a novel model, named double-reservoir echo state networks (DR-ESN), is proposed. DR-ESN is constructed by two reservoirs which are connected in series, thus the performance of abstracting the characteristics from the prediction task is improved. A sufficient condition is provided to ensure the stability of DR-ESN. The batch gradient method and ridge regression method are utilized to optimize the six parameters of DR-ESN and train the readouts, respectively. DR-ESN is verified by two different experiments, chaotic time series prediction and real-valued function time series prediction. The simulation results demonstrates that DR-ESN has a more precise result than leaky-ESN in predicting the time series.",

keywords = "Batch gradient descent, Double-reservoir, Echo state networks, Time-series prediction",

author = "Chong Liu and Huaguang Zhang and Xianshuang Yao and Kun Zhang",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 ; Conference date: 01-12-2016 Through 04-12-2016",

year = "2017",

month = mar,

day = "23",

doi = "10.1109/ICICIP.2016.7885901",

language = "英语",

series = "7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "196--202",

booktitle = "7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings",

address = "美国",

}

Liu, C, Zhang, H, Yao, X & Zhang, K 2017, Echo state networks with double-reservoir for time-series prediction. in 7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings., 7885901, 7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 196-202, 7th International Conference on Intelligent Control and Information Processing, ICICIP 2016, Siem Reap, Cambodia, 1/12/16. https://doi.org/10.1109/ICICIP.2016.7885901

Echo state networks with double-reservoir for time-series prediction. / Liu, Chong; Zhang, Huaguang; Yao, Xianshuang et al.
7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 196-202 7885901 (7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Echo state networks with double-reservoir for time-series prediction

AU - Liu, Chong

AU - Zhang, Huaguang

AU - Yao, Xianshuang

AU - Zhang, Kun

PY - 2017/3/23

Y1 - 2017/3/23

N2 - In this paper, a novel model, named double-reservoir echo state networks (DR-ESN), is proposed. DR-ESN is constructed by two reservoirs which are connected in series, thus the performance of abstracting the characteristics from the prediction task is improved. A sufficient condition is provided to ensure the stability of DR-ESN. The batch gradient method and ridge regression method are utilized to optimize the six parameters of DR-ESN and train the readouts, respectively. DR-ESN is verified by two different experiments, chaotic time series prediction and real-valued function time series prediction. The simulation results demonstrates that DR-ESN has a more precise result than leaky-ESN in predicting the time series.

AB - In this paper, a novel model, named double-reservoir echo state networks (DR-ESN), is proposed. DR-ESN is constructed by two reservoirs which are connected in series, thus the performance of abstracting the characteristics from the prediction task is improved. A sufficient condition is provided to ensure the stability of DR-ESN. The batch gradient method and ridge regression method are utilized to optimize the six parameters of DR-ESN and train the readouts, respectively. DR-ESN is verified by two different experiments, chaotic time series prediction and real-valued function time series prediction. The simulation results demonstrates that DR-ESN has a more precise result than leaky-ESN in predicting the time series.

KW - Batch gradient descent

KW - Double-reservoir

KW - Echo state networks

KW - Time-series prediction

UR - https://www.scopus.com/pages/publications/85018170795

U2 - 10.1109/ICICIP.2016.7885901

DO - 10.1109/ICICIP.2016.7885901

M3 - 会议稿件

AN - SCOPUS:85018170795

T3 - 7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings

SP - 196

EP - 202

BT - 7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 7th International Conference on Intelligent Control and Information Processing, ICICIP 2016

Y2 - 1 December 2016 through 4 December 2016

ER -

Liu C, Zhang H, Yao X, Zhang K. Echo state networks with double-reservoir for time-series prediction. In 7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 196-202. 7885901. (7th International Conference on Intelligent Control and Information Processing, ICICIP 2016 - Proceedings). doi: 10.1109/ICICIP.2016.7885901

Echo state networks with double-reservoir for time-series prediction

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this