user

Delayed Deep Echo State Network and Its Application on Time Series Prediction

被引：0
|
作者：
Bo Y.-C. [1 ]
Zhang X. [1 ]
Liu B. [1 ]
机构：
[1] College of Control Science and Engineering, China University of Petroleum (East China), Qingdao
来源：
Zidonghua Xuebao/Acta Automatica Sinica | 2020年 / 46卷 / 08期
基金：
中国国家自然科学基金;
关键词：
Artificial neural networks; Deep learning; Echo state network; Short-term memory capacity; Time series prediction;
D O I：
10.16383/j.aas.c180264
中图分类号：
学科分类号：
摘要：
To improve the prediction ability of echo state network (ESN) on time series problems, this paper proposes a delayed deep ESN (DDESN) constructing method. In this scheme, multiple sub-reservoirs are connected one by one in sequence, and time delay modules are inserted between every two adjacent sub-reservoirs. The DDESN can transfer a long-term memory task into a series of short-term memory tasks because of the existence of the delay links. It simplifies the solution to long-term dependent task and reduces the difficulty of building a reservoir. Experimental results show that the proposed DDESN has stronger short-term memory capacity, better robustness to randomly initialized parameters, and higher performance on solving time series tasks than a standard ESN. Copyright © 2020 Acta Automatica Sinica. All rights reserved.
引用
收藏
页码：1644 / 1653
页数：9
相关论文
共 28 条
  • [1] Qiao J F, Wang L, Yang C L, Gu K., Adaptive Levenberg-Marquardt algorithm based echo state network for chaotic time series prediction, IEEE Access, 6, pp. 10720-10732, (2018)
  • [2] Jaeger H, Haas H., Harnessing nonlinearity: Predicting chaotic systems and saving energy in wireless communication, Science, 304, 5667, pp. 78-80, (2004)
  • [3] Xu M L, Han M., Adaptive elastic echo state network for multivariate time series prediction, IEEE Transactions on Cybernetics, 46, 10, pp. 2173-2183, (2016)
  • [4] Han M, Xu M L., Laplacian echo state network for multivariate time series prediction, IEEE Transactions on Neural Networks & Learning Systems, 29, 1, pp. 238-244, (2018)
  • [5] Scardapane S, Wang D H, Panella M., A decentralized training algorithm for echo state networks in distributed big data applications, Neural Networks, 78, pp. 65-74, (2016)
  • [6] Bo Y C, Zhang X., Online adaptive dynamic programming based on echo state networks for dissolved oxygen control, Applied Soft Computing, 62, pp. 830-839, (2018)
  • [7] Wootton A J, Taylor S L, Day C R, Haycock P W., Optimizing echo state networks for static pattern recognition, Cognitive Computation, 9, 3, pp. 391-399, (2017)
  • [8] Trentin E, Scherer S, Schwenker F., Emotion recognition from speech signals via a probabilistic echo-state network, Pattern Recognition Letters, 66, pp. 4-12, (2015)
  • [9] Xu Mei-Ling, Han Min, Factor echo state network for multivariate chaotic time series prediction, Acta Automatica Sinica, 41, 5, pp. 1042-1046, (2015)
  • [10] Jaeger H., Reservoir riddles: Suggestions for echo state network research, Proceedings of the 2005 IEEE International Joint Conference on Neural Networks, pp. 1460-1462, (2005)
123