pith. sign in

arxiv: 2109.05364 · v1 · pith:PXXDE4UInew · submitted 2021-09-11 · 💻 cs.LG · physics.comp-ph

Structure-preserving Sparse Identification of Nonlinear Dynamics for Data-driven Modeling

classification 💻 cs.LG physics.comp-ph
keywords dynamicsdata-drivenidentificationlearningmodelingnonlinearsparsestructure
0
0 comments X
read the original abstract

Discovery of dynamical systems from data forms the foundation for data-driven modeling and recently, structure-preserving geometric perspectives have been shown to provide improved forecasting, stability, and physical realizability guarantees. We present here a unification of the Sparse Identification of Nonlinear Dynamics (SINDy) formalism with neural ordinary differential equations. The resulting framework allows learning of both "black-box" dynamics and learning of structure preserving bracket formalisms for both reversible and irreversible dynamics. We present a suite of benchmarks demonstrating effectiveness and structure preservation, including for chaotic systems.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.