Error-Driven Chained Multiple-Subnetwork Echo State Network for Time-Series Prediction

被引：10

作者：

Huang, Jian ^{[1
]}

Li, Yiran ^{[1
]}

Shardt, Yuri A. W. ^{[2
]}

Qiao, Liang ^{[1
]}

Shi, Mingrui ^{[1
]}

Yang, Xu ^{[1
]}

机构：

[1] Univ Sci & Technol Beijing, Sch Automat & Elect Engn, Key Lab Knowledge Automat Ind Proc, Minist Educ, Beijing 100083, Peoples R China

[2] Tech Univ Ilmenau, Dept Automat Engn, D-98684 Ilmenau, Germany

来源：

IEEE SENSORS JOURNAL | 2022年 / 22卷 / 20期

基金：

北京市自然科学基金;

关键词：

Reservoirs; Predictive models; Time series analysis; Topology; Optimization; Computational modeling; Training; Echo state network (ESN); error-driven chain topology; multiple subnetworks; time-series prediction; CYCLE RESERVOIR NETWORK; OPTIMIZATION; MULTISTEP; PROPERTY; DESIGN;

D O I：

10.1109/JSEN.2022.3200069

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Hybrid echo state networks (ESNs), a type of modified ESN, have been developed to improve the prediction accuracy of ESNs. However, they have been criticized for their computational complexity, which makes it difficult to use them directly in industrial applications. In this article, an error-driven chained multiple-subnetwork ESN (CESN) is proposed to build a simple structured hybrid network and improve its prediction accuracy. For this reason, a chain topology is generated to gradually reduce the residual error, while each subnetwork is trained separately. The weight matrix for each subnetwork does not need to be optimized, which reduces the computational cost. Meanwhile, the optimal number of subnetworks is determined on the basis of a given application. The efficiency of the proposed CESN is tested on a Santa Fe Laser and a public building dataset. Compared with ESN, 70% of the test data have been optimized by CESN for the public building dataset.

引用

页码：19533 / 19542

页数：10

共 50 条

[21] Laplacian Echo State Network for Multivariate Time Series Prediction
Han, Min
Xu, Meiling
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2018, 29 (01) : 238 - 244
[22] Echo state network implementation for chaotic time series prediction
de la Fraga, Luis Gerardo
Ovilla-Martinez, Brisbane
Tlelo-Cuautle, Esteban
MICROPROCESSORS AND MICROSYSTEMS, 2023, 103
[23] Dynamical regularized echo state network for time series prediction
Cuili Yang
Junfei Qiao
Lei Wang
Xinxin Zhu
Neural Computing and Applications, 2019, 31 : 6781 - 6794
[24] Echo State Network With Probabilistic Regularization for Time Series Prediction
Xiufang Chen
Mei Liu
Shuai Li
IEEE/CAA Journal of Automatica Sinica, 2023, 10 (08) : 1743 - 1753
[25] Chaotic time series prediction by noisy echo state network
Shinozaki, Aren
Miyano, Takaya
Horio, Yoshihiko
IEICE NONLINEAR THEORY AND ITS APPLICATIONS, 2020, 11 (04): : 466 - 479
[26] An Echo State Network With Improved Topology for Time Series Prediction
Li, Xin
Bi, Fengrong
Yang, Xiao
Bi, Xiaoyang
IEEE SENSORS JOURNAL, 2022, 22 (06) : 5869 - 5878
[27] A Modified Echo State Network in Chaotic Time Series Prediction
Li Dingyuan
Liu Fu
Qiao Junfei
Li Rong
2017 29TH CHINESE CONTROL AND DECISION CONFERENCE (CCDC), 2017, : 4350 - 4353
[28] Minimum mean squared error time series classification using an echo state network prediction model
Skowronski, Mark D.
Harris, John G.
2006 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS, VOLS 1-11, PROCEEDINGS, 2006, : 3153 - +
[29] Small-World Echo State Networks for Nonlinear Time-Series Prediction
Mo, Shu
Hu, Kai
Li, Weibing
Pan, Yongping
NEURAL INFORMATION PROCESSING, ICONIP 2023, PT II, 2024, 14448 : 550 - 560
[30] Support vector echo-state machine for chaotic time-series prediction
Shi, Zhiwei
Han, Min
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2007, 18 (02): : 359 - 372

← 1 2 3 4 5 →