Data-driven inference of complex system dynamics: A mini-review
- 24 March 2023
- journal article
- review article
- Published by IOP Publishing in Europhysics Letters
- Vol. 142 (1), 11001
- https://doi.org/10.1209/0295-5075/acc3bf
Abstract
Our ability to observe the network topology and nodes' behaviors of complex systems has significantly advanced in the past decade, giving rise to a new and fast-developing frontier -inferring the underlying dynamical mechanisms of complex systems from the observation data. Here we explain the rationale of data-driven dynamics inference and review the recent progress in this emerging field. Specifically, we classify the existing methods of dynamics inference into three categories, and describe their key ideas, representative applications and limitations. We also discuss the remaining challenges that are worth the future effort. perspective Copyright (c) 2023 EPLAThis publication has 50 references indexed in Scilit:
- Quantifying causal emergence shows that macro can beat microProceedings of the National Academy of Sciences of the United States of America, 2013
- Universality in network dynamicsNature Physics, 2013
- Microscopic, mesoscopic and macroscopic descriptions of complex systemsProbabilistic Engineering Mechanics, 2011
- Eureqa: software reviewGenetic Programming and Evolvable Machines, 2010
- Perspectives on system identificationAnnual Reviews in Control, 2010
- Distilling Free-Form Natural Laws from Experimental DataScience, 2009
- Network structure of cerebral cortex shapes functional connectivity on multiple time scalesProceedings of the National Academy of Sciences of the United States of America, 2007
- Genetic programming as a means for programming computers by natural selectionStatistics and Computing, 1994
- Neural networks and physical systems with emergent collective computational abilities.Proceedings of the National Academy of Sciences of the United States of America, 1982
- Editorial: Effective construction of linear state-variable models from input/output functionsat – Automatisierungstechnik, 1966