Large language models for automatic equation discovery of nonlinear dynamics
Physics of Fluids · 01.09.2024 · 36 (9)
DOI: 10.1063/5.0224297 ↗Kurz gefasst
Fachartikel in der Zeitschrift Physics of Fluids, online erschienen am 10.09.2024. Verfasst von Mengge Du und 4 weiteren Autor:innen. Die Arbeit wurde bisher 32-mal zitiert (Crossref) und nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Center for Applied Mathematics Shenzhen, Shenzhen Key Laboratory of Natural Gas Hydrates, SUSTech-Qingdao New Energy Technology Research Institute.
Abstract (Originalsprache)
Equation discovery aims to directly extract physical laws from data and has emerged as a pivotal research domain in nonlinear systems. Previous methods based on symbolic mathematics have achieved substantial advancements, but often require handcrafted representation rules and complex optimization algorithms. In this paper, we introduce a novel framework that utilizes natural language-based prompts to guide large language models (LLMs) in automatically extracting governing equations from data. Specifically, we first utilize the generation capability of LLMs to generate diverse candidate equations in string form and then evaluate the generated equations based on observations. The best equations are preserved and further refined iteratively using the reasoning capacity of LLMs. We propose two alternately iterated strategies to collaboratively optimize the generated equations. The first strategy uses LLMs as a black-box optimizer to achieve equation self-improvement based on historical samples and their performance. The second strategy instructs LLMs to perform evolutionary operations for a global search. Experiments are conducted on various nonlinear systems described by partial differential equations, including the Burgers equation, the Chafee–Infante equation, and the Navier–Stokes equation. The results demonstrate that our framework can discover correct equations that reveal the underlying physical laws. Further comparisons with state-of-the-art models on extensive ordinary differential equations showcase that the equations discovered by our framework possess physical meaning and better generalization capability on unseen data.
Förderung
- National Center for Applied Mathematics Shenzhen
- Shenzhen Key Laboratory of Natural Gas Hydrates
- SUSTech-Qingdao New Energy Technology Research Institute
- China Meteorological Administration Climate Change Special Program
- National Natural Science Foundation of China
Zitieren
APA
Du, M., Chen, Y., Wang, Z., Nie, L., & Zhang, D. (2024). Large language models for automatic equation discovery of nonlinear dynamics. Physics of Fluids, 36(9). https://doi.org/10.1063/5.0224297
Vancouver
Du M, Chen Y, Wang Z, Nie L, Zhang D. Large language models for automatic equation discovery of nonlinear dynamics. Physics of Fluids. 2024;36(9). doi:10.1063/5.0224297
BibTeX
@article{Du2024,
author = {Du, Mengge and Chen, Yuntian and Wang, Zhongzheng and Nie, Longfeng and Zhang, Dongxiao},
title = {Large language models for automatic equation discovery of nonlinear dynamics},
journal = {Physics of Fluids},
year = {2024},
volume = {36},
number = {9},
doi = {10.1063/5.0224297}
}
EUVIMED