Learning Dynamical Systems and Bifurcation via Group Sparsity

Abstract

Learning governing equations from a family of data sets which share the same physical laws but differ in bifurcation parameters is challenging. This is due, in part, to the wide range of phenomena that could be represented in the data sets as well as the range of parameter values. On the other hand, it is common to assume only a small number of candidate functions contribute to the observed dynamics. Based on these observations, we propose a group-sparse penalized method for model selection and parameter estimation for such data. We also provide convergence guarantees for our proposed numerical scheme. Various numerical experiments including the 1D logistic equation, the 3D Lorenz sampled from different bifurcation regions, and a switching system provide numerical validation for our method and suggest potential applications to applied dynamical systems.

16 pages, 18 figures

References in corpus (4)

• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Sparsity and Incoherence in Compressive Sampling
• Simultaneous Orthogonal Matching Pursuit With Noise Stabilization: Theoretical Analysis
• Nonparametric Simultaneous Sparse Recovery: an Application to Source Localization

Cited by in corpus (6)

• Learning physically consistent mathematical models from data using group sparsity
• NeuPDE: Neural Network Based Ordinary and Partial Differential Equations for Modeling Time-Dependent Data
• Deeptime: a Python library for machine learning dynamical models from time series data
• On the Convergence of the SINDy Algorithm
• Extracting structured dynamical systems using sparse optimization with very few samples
• Recovery guarantees for polynomial approximation from dependent data with outliers