CTSR: Cartesian tensor-based sparse regression for data-driven discovery of high-dimensional invariant governing equations

2025/04/10 by Zhang, Boqian, Lei, Juanmian, Sun, Guoyou +2 · 2 citations
#FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.2504.07618

Abstract

Accurate and concise governing equations are crucial for understanding system dynamics. Recently, data-driven methods such as sparse regression have been employed to automatically uncover governing equations from data, representing a significant shift from traditional first-principles modeling. However, most existing methods focus on scalar equations, limiting their applicability to simple, low-dimensional scenarios, and failing to ensure rotation and reflection invariance without incurring significant computational cost or requiring additional prior knowledge. This paper proposes a Cartesian tensor-based sparse regression (CTSR) technique to accurately and efficiently uncover complex, high-dimensional governing equations while ensuring invariance. Evaluations on two two-dimensional (2D) and two three-dimensional (3D) test cases demonstrate that the proposed method achieves superior accuracy and efficiency compared to the conventional technique.

Citations

• Constraining Genetic Symbolic Regression via Semantic Backpropagation
• An invariance constrained deep learning network for PDE discovery
• Nonlinear parametric models of viscoelastic fluid flows
• Discover governing differential equations from evolving systems
• PDE-LEARN: Using Deep Learning to Discover Partial Differential Equations from Noisy, Limited Data
• Interpretable scientific discovery with symbolic regression: a review
• Dimensional homogeneity constrained gene expression programming for discovering governing equations
• Deep Learning and Symbolic Regression for Discovering Parametric Equations
• Robust learning from noisy, incomplete, high-dimensional experimental data via physically constrained symbolic regression
• Physics-informed learning of governing equations from scarce data
• SINDy-PI: A Robust Algorithm for Parallel Implicit Sparse Identification of Nonlinear Dynamics
• Integration of Neural Network-Based Symbolic Regression in Deep Learning for Scientific Discovery
• Using Noisy or Incomplete Data to Discover Models of Spatiotemporal Dynamics
• Deep symbolic regression: Recovering mathematical expressions from data via risk-seeking policy gradients
• Robust and optimal sparse regression for nonlinear PDE models
• AI Feynman: a Physics-Inspired Method for Symbolic Regression
• Improving Model-based Genetic Programming for Symbolic Regression of Small Expressions
• Learning Equations for Extrapolation and Control
• Extrapolation and learning equations
• Data-driven discovery of partial differential equations
• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Genetic programming as a means for programming computers by natural selection
• A new constitutive equation derived from network theory
• On the formulation of rheological equations of state
• Stabilization of an open-source finite-volume solver for viscoelastic fluid flows
• A simple constitutive equation for polymer fluids based on the concept of deformation-dependent tensorial mobility

Cited by

• Symbolic identification of tensor equations in multidimensional physical fields

Related

• Sparse and Low-Rank High-Order Tensor Regression via Parallel Proximal Method
• Sparse Regression and Adaptive Feature Generation for the Discovery of Dynamical Systems
• Provable Sparse Tensor Decomposition
• Discovery of differential equations using sparse state and parameter regression
• Boosted Sparse and Low-Rank Tensor Regression