Hierarchical plasticity echo state network for chaotic time series prediction

被引：0

作者：

Na X.-D. ^{[1
]}

Wang J.-N. ^{[1
]}

Liu M.-R. ^{[1
]}

Ren W.-J. ^{[2
]}

Han M. ^{[3
,4
]}

机构：

[1] Faculty of Electronic Information and Electrical Engineering, Dalian University of Technology, Dalian

[2] College of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin

[3] Key Laboratory of Intelligent Control and Optimization for Industrial Equipment, Ministry of Education, Dalian University of Technology, Dalian

[4] Professional Technology Innovation Center of Distributed Control for Industrial Equipment of Liaoning Province, Dalian University of Technology, Dalian

来源：

Kongzhi yu Juece/Control and Decision | 2023年 / 38卷 / 01期

关键词：

chaotic time series prediction; echo state network; hierarchical strategy; neural network; neuronal intrinsic plasticity; pre-training;

D O I：

10.13195/j.kzyjc.2021.0773

中图分类号：

学科分类号：

摘要：

To improve the ability of echo state network for feature extraction and prediction on chaotic time series, a hierarchical plasticity echo state network (HPESN) model is proposed. In this model, multiple reservoirs are connected in sequence, and the ability of nonlinear multi-scale dynamic feature extraction is enhanced through layer-by-layer feature transformation. Meanwhile, the intrinsic plasticity mechanism in neuroscience is introduced to simulate the firing rate distribution of real biological neurons, and the reservoir is pre-trained with the goal of maximizing neuronal information transmission. The HPESN not only increases the capacity of the model and reduces the instability caused by random projection, but also provides a new idea for understanding the representation, processing, memory and storage operations of the reservoir. The simulation results show that compared with other seven modified echo state network models, the proposed HPESN model achieves the best prediction accuracy in the chaotic time series prediction task composed of synthetic data and real-world data. © 2023 Northeast University. All rights reserved.

引用

页码：133 / 142

页数：9

共 32 条

[1]

Tang Y, Kurths J, Lin W, Et al., Introduction to focus issue: When machine learning meets complex systems: Networks, chaos, and nonlinear dynamics, Chaos: An Interdisciplinary Journal of Nonlinear Science, 30, 6, (2020)

[2]

Mudelsee M., Trend analysis of climate time series: A review of methods, Earth-Science Reviews, 190, pp. 310-322, (2019)

[3]

Wang D, Wang P, Shi J Z., An interval prediction algorithm for chaos time series with controllable error rate, Control and Decision, 34, 5, pp. 956-964, (2019)

[4]

Li D C, Han M, Wang J., Chaotic time series prediction based on a novel robust echo state network, IEEE Transactions on Neural Networks and Learning Systems, 23, 5, pp. 787-799, (2012)

[5]

Jaeger H, Haas H., Harnessing nonlinearity: Predicting chaotic systems and saving energy in wireless communication, Science, 304, 5667, pp. 78-80, (2004)

[6]

Jaeger H, Lukosevicius M, Popovici D, Et al., Optimization and applications of echo state networks with leaky-integrator neurons, Neural Networks, 20, 3, pp. 335-352, (2007)

[7]

Lukosevicius M, Jaeger H., Reservoir computing approaches to recurrent neural network training, Computer Science Review, 3, 3, pp. 127-149, (2009)

[8]

Lu Z X, Hunt B R, Ott E., Attractor reconstruction by machine learning, Chaos: An Interdisciplinary Journal of Nonlinear Science, 28, 6, (2018)

[9]

Grigoryeva L, Ortega J P., Echo state networks are universal, Neural Networks, 108, pp. 495-508, (2018)

[10]

Wang J Z, Niu T, Lu H Y, Et al., A novel framework of reservoir computing for deterministic and probabilistic wind power forecasting, IEEE Transactions on Sustainable Energy, 11, 1, pp. 337-349, (2020)

← 1 2 3 4 →