Model-free prediction of chaotic dynamics with parameter-aware reservoir computing
Jianmin Guo(郭建敏)1,2, Yao Du(杜瑶)1, Haibo Luo(罗海波)1, Xuan Wang(王晅)1, Yizhen Yu(于一真)1,†, and Xingang Wang(王新刚)1
1 School of Physics and Information Technology, Shaanxi Normal University, Xi'an 710062, China; 2 College of Physics and Electronic Information Engineering, Qinghai Normal University, Xining 810008, China
Abstract Model-free, data-driven prediction of chaotic motions is a long-standing challenge in nonlinear science. Stimulated by the recent progress in machine learning, considerable attention has been given to the inference of chaos by the technique of reservoir computing (RC). In particular, by incorporating a parameter-control channel into the standard RC, it is demonstrated that the machine is able to not only replicate the dynamics of the training states, but also infer new dynamics not included in the training set. The new machine-learning scheme, termed parameter-aware RC, opens up new avenues for data-based analysis of chaotic systems, and holds promise for predicting and controlling many real-world complex systems. Here, using typical chaotic systems as examples, we give a comprehensive introduction to this powerful machine-learning technique, including the algorithm, the implementation, the performance, and the open questions calling for further studies.
Fund: Project supported by the National Natural Science Foundation of China (Grant No. 12275165). XGW was also supported by the Fundamental Research Funds for the Central Universities (Grant No. GK202202003).
Jianmin Guo(郭建敏), Yao Du(杜瑶), Haibo Luo(罗海波), Xuan Wang(王晅), Yizhen Yu(于一真), and Xingang Wang(王新刚) Model-free prediction of chaotic dynamics with parameter-aware reservoir computing 2025 Chin. Phys. B 34 040505
[1] Ott E 2022 Chaos in Dynamical Systems 2nd Edn. (Cambridge: Cambridge University Press) [2] Strogatz S H 2015 Nonlinear Dynamics and Chaos: With Applications to Physics, Biology, Chemistry, and Engineering (CRC Press) [3] Wang W X, Yang R, Lai Y C, Kovanis V and Grebogi C 2011 Phys. Rev. Lett. 106 154101 [4] Lai Y C 2021 Chaos 31 082101 [5] MaassW, Natschlager T and Markram H 2002 Neural Comput. 14 2531 [6] Jaeger H and Haas H 2004 Science 304 78 [7] Pathak J, Lu Z, Hunt B, Girvan M and Ott E 2017 Chaos 27 121102 [8] Lu Z, Hunt B R and Ott E 2018 Chaos 28 061104 [9] Weng T, Yang H, Gu C, Zhang J and Small M 2019 Phys. Rev. E 99 042203 [10] Fan H, Jiang J, Zhang C, Wang X G and Lai Y C 2020 Phys. Rev. Res. 2 012080 [11] Guo Y L, Zhang H, Wang L, Fan H, Xiao J H and Wang X G 2021 Chaos 31 011104 [12] Wang L, Fan H, Xiao J, Lan Y and Wang X G 2022 Phys. Rev. E 105 L052201 [13] Kong L W, Weng Y, Glaz B, Haile M and Lai Y C 2023 Chaos 33 033111 [14] Du Y, Li Q, Fan H, Zhan M, Xiao J and Wang X G 2024 Phys. Rev. Res. 6 013181 [15] Lin Z, Lu Z, Di Z and Tang Y 2024 Nat. Commun. 15 6584 [16] Panahi S and Lai Y C 2024 Chaos 34 051501 [17] Lukosevicius M and Jaeger H 2009 Comput. Sci. Rev. 3 127 [18] Tanaka G, Yamane T, Heroux J B, Nakane R, Kanazawa N, Takeda S, Numata H, Nakano D and Hirose A 2019 Neural Networks 115 100 [19] Yan M, Huang C, Bienstman P, Tino P, Lin W and Sun J 2024 Nat. Commun. 15 2056 [20] Pathak J, Hunt B, Girvan M, Lu Z and Ott E 2018 Phys. Rev. Lett. 120 024102 [21] Szunyogh I, Arcomano T, Pathak J, Wikner A, Hunt B and Ott E 2020 Geophys. Res. Lett. 47 e2020GL087776 [22] Klos C, Kossio Y F K, Goedeke S, Gilra A and Memmesheimer R M 2020 Phys. Rev. Lett. 125 088103 [23] Zhao H 2021 Sci. China-Phys. Mech. Astron. 64 270511 [24] Kong L W, Fan H, Grebogi C and Lai Y C 2021 Phys. Rev. Res. 3 013090 [25] Fan H, Kong L W, Lai Y C and Wang X G 2021 Phys. Rev. Res. 3 023237 [26] Xiao R, Kong LW, Sun Z K and Lai Y C 2021 Phys. Rev. E 104 014205 [27] Zhang H, Fan H,Wang L andWang X G 2021 Phys. Rev. E 104 024205 [28] Kim J Z, Lu Z, Nozari E, Pappas G J and Bassett D S 2021 Nat. Mach. Intell. 3 316 [29] Zhang H, Fan H, Du Y,Wang L andWang X G 2022 Chaos 32 083136 [30] Fan H, Wang L, Du Y, Wang Y, Xiao J and Wang X G 2022 Phys. Rev. Res. 4 013137 [31] Luo H, Du Y, Fan H, Wang X, Guo J and Wang X G 2024 Phys. Rev. E 109 024210 [32] Sisodia D and Jalan S 2024 Phys. Rev. E 110 034211 [33] Zhai Z M, Moradi M, Blaz B, Haile M and Lai Y C 2024 Phys. Rev. Res. 6 013196 [34] Kong L W, Brewer G A and Lai Y C 2024 Nat. Commun. 15 4840 [35] Du Y, Luo H, Guo J, Xiao J, and Wang X G 2024 arXiv:2409.16719nlin.CD] [36] Lorenz E N 1996 Predictability: A problem partly solved, in Proceedings Seminar on Predictability (ECMWF) Vol. 1 [37] Gauthier D J, Bollt E, Griffith A and Wendson A S B 2021 Nat. Commun. 12 5564
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.
