Data-driven discovery of partial differential equations
By Samuel H. Rudy, Steven L. Brunton, Joshua L. Proctor, J. Nathan Kutz. 2016
1,709 citations recorded by the source corpus.
Researchers propose sparse regression for identifying governing partial differential equations for spatiotemporal systems. We propose a sparse regression method capable of discovering the governing partial differential equation(s) of a given system by time series measurements in the spatial domain. The regression framework relies on sparsity-promoting techniques to select the nonlinear and partial derivative terms of the governing equations that most accurately represent the data, bypassing a combinatorially large search through all possible candidate models. The method balances model complexity and regression accuracy by selecting a parsimonious model via Pareto analysis. Time series measurements can be made in an Eulerian framework, where the sensors are fixed spatially, or in a Lagrangian framework, where the sensors move with the dynamics. The method is computationally efficient, robust, and demonstrated to work on a variety of canonical problems spanning a number of scientific domains including Navier-Stokes, the quantum harmonic oscillator, and the diffusion equation. Moreover, the method is capable of disambiguating between potentially nonunique dynamical terms by using multiple time series taken with different initial data. Thus, for a traveling wave, the method can distinguish between a linear wave equation and the Korteweg–de Vries equation, for instance. The method provides a promising new technique for discovering governing equations and physical laws in parameterized spatiotemporal systems, where first-principles derivations are intractable.
Subfields: Numerical Analysis, Statistical and Nonlinear Physics, Statistics and Probability, Numerical Methods & Scientific Computing, AI for Physical Sciences
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Numerical Analysis
- Steven L. Brunton
- Maizar Raissi
- G. Em Karniadakis
- J. Nathan Kutz
- Paris Perdikaris
- Bingni W. Brunton
- Zhao Chen
- Yang Liu
In Statistical and Nonlinear Physics
- G. Em Karniadakis
- Steven L. Brunton
- Maizar Raissi
- J. Nathan Kutz
- Paris Perdikaris
- Lu Lu
- Xuhui Meng
- Yang Liu
In Statistics and Probability
- G. Em Karniadakis
- Steven L. Brunton
- J. Nathan Kutz
- Paris Perdikaris
- Maizar Raissi
- Lu Lu
- Xuhui Meng
- Dongbin Xiu
In Numerical Methods & Scientific Computing
- Maizar Raissi
- G. Em Karniadakis
- Paris Perdikaris
- Jens Berg
- Kaj Nyström
- Dongkun Zhang
- Steven L. Brunton
- Liu Yang
In AI for Physical Sciences
- G. Em Karniadakis
- Maizar Raissi
- Paris Perdikaris
- Steven L. Brunton
- J. Nathan Kutz
- Lu Lu
- Jens Berg
- Kaj Nyström
Citations in the corpus
- Interpretable Machine Learning: Fundamental Principles and 10 Grand Challenges (2021, methodology): “Several additional works [167, 43, 259] used sparse regression and dynamic mode decomposition to discover the governing equa-”
- fPINNs: Fractional Physics-Informed Neural Networks (2018, background): “In this paper, we focus on the identification of the parameters of fractional PDEs whose overall form is known but some coefficients and most importantly the operators themselves are not known.”
- Deep Hidden Physics Models: Deep Learning of Nonlinear Partial Differential Equations (2018, methodology): “To generate a training dataset for this problem we follow the exact same instructions as the ones provided in (Kutz et al., 2016; Rudy et al., 2017).”
- Multistep Neural Networks for Data-driven Discovery of Nonlinear Dynamical Systems (2018, methodology): “In particular, following the problem setup presented in [23] and [24], we simulate the Navier-Stokes equations describing the two-dimensional fluid flow past a circular cylinder at Reynolds number 100 using the Immersed Boundary Projection Method [25, 26].”
- Hidden physics models: Machine learning of nonlinear partial differential equations (2017, methodology): “The original data-set proposed in [2] contains 501 time snapshots of a solution to the nonlinear Schrödinger equation with a Gaussian initial condition.”
- Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equations (2017, background): “In this second part of our study, we shift our attention to the problem of data-driven discovery of partial differential equations [7, 8, 9].”
- Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations (2017, background): “The first problem is that of predictive inference, filtering and smoothing, or data driven solutions of partial differential equations [9, 5] which states: given fixed model parameters λ what can be said about the unknown hidden state u(t, x) of the system? The second problem is that of learning, sy”
- PDE-Net: Learning PDEs from Data (2017, methodology): “Recently, Brunton et al. (2016), Schaeffer (2017), Rudy et al. (2017) and Wu & Zhang (2017) propose an alternative approach using sparse regression.”
- Machine learning of linear differential equations using Gaussian processes (2017, background): “However, regularization still remains an important factor even in cases where data is abundant as seen in the recently growing literature on discovering ordinary and partial differential equations from data using sparse regression techniques [7,41].”
- Machine learning for molecular and materials science (2018, background): “Automatic discovery of scientific laws and principles 99-100 by inspection of the weights of trained ML systems is a potentially transformational development in science.”
- SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics (2020, methodology): “Figure 5 shows the results of SINDy-PI and PDE-FIND [29] for different values of g 0 .”
- Model selection for dynamical systems via sparse regression and information criteria (2017, background): “to handle rational function nonlinearities [4], partial differential equations [2], highly corrupted data [27], and to build Galerkin regression models in fluids [28].”
- Constrained sparse Galerkin regression (2016, methodology): “However, the extensions of SINDy to identify partial differential equations (Rudy et al. 2017; Schaeffer 2017) and biological regulatory networks (Mangan et al. 2016) highlight the flexibility of this simple regression framework.”
- Lift & Learn: Physics-informed machine learning for large-scale nonlinear dynamical systems (2019, methodology): “Techniques from compressed sensing are also used in [12, 13] to fit terms of governing equations from a dictionary of possible nonlinear terms.”