Detecting critical transitions before they occur is challenging, especially for complex dynamical systems. While some early-warning indicators have been suggested to capture the phenomenon of slowing down in the system's response near critical transitions, their applicability to real systems is yet limited. In this paper, we propose the concept of predictability based on machine learning methods, which leads to an alternative early-warning indicator. The predictability metric takes a black-box approach and assesses the impact of uncertainties itself in identifying abrupt transitions in time series. We have applied the proposed metric to the time series generated from different systems, including an ecological model and an electric power system. We show that the predictability changes noticeably before critical transitions occur, while other general indicators such as variance and autocorrelation fail to make any notable signals.
Citation: Jaesung Choi, Pilwon Kim. Early warning for critical transitions using machine-based predictability[J]. AIMS Mathematics, 2022, 7(11): 20313-20327. doi: 10.3934/math.20221112
Detecting critical transitions before they occur is challenging, especially for complex dynamical systems. While some early-warning indicators have been suggested to capture the phenomenon of slowing down in the system's response near critical transitions, their applicability to real systems is yet limited. In this paper, we propose the concept of predictability based on machine learning methods, which leads to an alternative early-warning indicator. The predictability metric takes a black-box approach and assesses the impact of uncertainties itself in identifying abrupt transitions in time series. We have applied the proposed metric to the time series generated from different systems, including an ecological model and an electric power system. We show that the predictability changes noticeably before critical transitions occur, while other general indicators such as variance and autocorrelation fail to make any notable signals.
| [1] |
C. N. Anderson, C.-h. Hsieh, S. A. Sandin, R. Hewitt, A. Hollowed, J. Beddington, et al., Why fishing magnifies fluctuations in fish abundance, Nature, 452 (2008), 835–839. https://doi.org/10.1038/nature06851 doi: 10.1038/nature06851
|
| [2] |
V. Dakos, S. R. Carpenter, W. A. Brock, A. M. Ellison, V. Guttal, A. R. Ives, et al., Methods for detecting early warnings of critical transitions in time series illustrated using simulated ecological data, PloS one, 7 (2012), e41010. https://doi.org/10.1371/journal.pone.0041010 doi: 10.1371/journal.pone.0041010
|
| [3] |
I. Dobson, H.-D. Chiang, Towards a theory of voltage collapse in electric power systems, Syst. Control Lett., 13 (1989), 253–262. https://doi.org/10.1016/0167-6911(89)90072-8 doi: 10.1016/0167-6911(89)90072-8
|
| [4] |
H. Fan, L.-W. Kong, Y.-C. Lai, X. Wang, Anticipating synchronization with machine learning, Phys. Rev. Res., 3 (2021), 023237. https://doi.org/10.1103/PhysRevResearch.3.023237 doi: 10.1103/PhysRevResearch.3.023237
|
| [5] |
A. S. Gsell, U. Scharfenberger, D. Özkundakci, A. Walters, L.-A. Hansson, A. B. Janssen, et al., Evaluating early-warning indicators of critical transitions in natural aquatic ecosystems, Proceedings of the National Academy of Sciences, 113 (2016), E8089–E8095. https://doi.org/10.1073/pnas.1608242113 doi: 10.1073/pnas.1608242113
|
| [6] |
A. Haluszczynski, J. Aumeier, J. Herteux, C. Räth, Reducing network size and improving prediction stability of reservoir computing, Chaos: An Interdisciplinary Journal of Nonlinear Science, 30 (2020), 063136. https://doi.org/10.1063/5.0006869 doi: 10.1063/5.0006869
|
| [7] |
A. Haluszczynski, C. Räth, Good and bad predictions: Assessing and improving the replication of chaotic attractors by means of reservoir computing, Chaos: An Interdisciplinary Journal of Nonlinear Science, 29 (2019), 103143. https://doi.org/10.1063/1.5118725 doi: 10.1063/1.5118725
|
| [8] |
L. Huang, Q. Chen, Y.-C. Lai, L. M. Pecora, Generic behavior of master-stability functions in coupled nonlinear dynamical systems, Phys. Rev. E, 80 (2009), 036204. https://doi.org/10.1103/PhysRevE.80.036204 doi: 10.1103/PhysRevE.80.036204
|
| [9] |
A. R. Ives, Measuring resilience in stochastic systems, Ecol. Monogr., 65 (1995), 217–233. https://doi.org/10.2307/2937138 doi: 10.2307/2937138
|
| [10] | H. Jaeger, The "echo state" approach to analysing and training recurrent neural networks-with an erratum note, Bonn, Germany: German National Research Center for Information Technology GMD Technical Report, 148 (2001), 13. |
| [11] |
H. Jaeger, H. Haas, Harnessing nonlinearity: Predicting chaotic systems and saving energy in wireless communication, science, 304 (2004), 78–80. https://doi.org/10.1126/science.1091277 doi: 10.1126/science.1091277
|
| [12] |
B. Kaszás, U. Feudel, T. Tél, Tipping phenomena in typical dynamical systems subjected to parameter drift, Scientific reports, 9 (2019), 1–12. https://doi.org/10.1038/s41598-019-44863-3 doi: 10.1038/s41598-019-44863-3
|
| [13] |
L.-W. Kong, H.-W. Fan, C. Grebogi, Y.-C. Lai, Machine learning prediction of critical transition and system collapse, Phys. Rev. Res., 3 (2021), 013090. https://doi.org/10.1103/PhysRevResearch.3.013090 doi: 10.1103/PhysRevResearch.3.013090
|
| [14] |
G. Kou, Y. Xu, Y. Peng, F. Shen, Y. Chen, K. Chang, et al., Bankruptcy prediction for smes using transactional data and two-stage multiobjective feature selection, Decis. Support Syst., 140 (2021), 113429. https://doi.org/10.1016/j.dss.2020.113429 doi: 10.1016/j.dss.2020.113429
|
| [15] |
S. J. Lade, T. Gross, Early warning signals for critical transitions: a generalized modeling approach, PLoS comput. biol., 8 (2012), e1002360. https://doi.org/10.1371/journal.pcbi.1002360 doi: 10.1371/journal.pcbi.1002360
|
| [16] |
T. M. Lenton, Early warning of climate tipping points, Nat. clim. change, 1 (2011), 201–209. https://doi.org/10.1038/nclimate1143 doi: 10.1038/nclimate1143
|
| [17] |
Z. Lu, J. Pathak, B. Hunt, M. Girvan, R. Brockett, E. Ott, Reservoir observers: Model-free inference of unmeasured variables in chaotic systems, Chaos: An Interdisciplinary Journal of Nonlinear Science, 27 (2017), 041102. https://doi.org/10.1063/1.4979665 doi: 10.1063/1.4979665
|
| [18] |
K. McCann, P. Yodzis, Nonlinear dynamics and population disappearances, The American Naturalist, 144 (1994), 873–879. https://doi.org/10.1086/285714 doi: 10.1086/285714
|
| [19] |
I. Noy-Meir, Stability of grazing systems: An application of predator-prey graphs, J. Ecol., 63 (1975), 459–481. https://doi.org/10.2307/2258730 doi: 10.2307/2258730
|
| [20] |
M. Scheffer, J. Bascompte, W. A. Brock, V. Brovkin, S. R. Carpenter, V. Dakos, et al., Early-warning signals for critical transitions, Nature, 461 (2009), 53–59. https://doi.org/10.1038/nature08227 doi: 10.1038/nature08227
|
| [21] |
M. Scheffer, S. R. Carpenter, T. M. Lenton, J. Bascompte, W. Brock, V. Dakos, et al., Anticipating critical transitions, science, 338 (2012), 344–348. https://doi.org/10.1126/science.1225244 doi: 10.1126/science.1225244
|
| [22] | Y. Uwatet, M. Schule, T. Ott, Y. Nishiot, Echo state network with chaos noise for time series prediction, in International Symposium on Nonlinear Theory and its Applications (NOLTA), Okinawa, Japan, 16–19 November 2020, (2020), 274. |
| [23] |
H. O. Wang, E. H. Abed, A. M. Hamdan, Bifurcations, chaos, and crises in voltage collapse of a model power system, IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications, 41 (1994), 294–302. https://doi.org/10.1109/81.285684 doi: 10.1109/81.285684
|
Figures(8)
Jaesung Choi, Pilwon Kim. Early warning for critical transitions using machine-based predictability[J]. AIMS Mathematics, 2022, 7(11): 20313-20327. doi: 10.3934/math.20221112
DownLoad: