DeepMoD: Deep learning for Model Discovery in noisy data

Abstract

We introduce DeepMoD, a Deep learning based Model Discovery algorithm. DeepMoD discovers the partial differential equation underlying a spatio-temporal data set using sparse regression on a library of possible functions and their derivatives. A neural network approximates the data and constructs the function library, but it also performs the sparse regression. This construction makes it extremely robust to noise, applicable to small data sets, and, contrary to other deep learning methods, does not require a training set. We benchmark our approach on several physical problems such as the Burgers', Korteweg-de Vries and Keller-Segel equations, and find that it requires as few as $\mathcal{O}(10^2)$ samples and works at noise levels up to $75\%$. Motivated by these results, we apply DeepMoD directly on noisy experimental time-series data from a gel electrophoresis experiment and find that it discovers the advection-diffusion equation describing this system.

References in corpus (10)

• Theory-guided Data Science: A New Paradigm for Scientific Discovery from Data
• Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations
• PDE-Net 2.0: Learning PDEs from Data with A Numeric-Symbolic Hybrid Deep Network
• Deep Hidden Physics Models: Deep Learning of Nonlinear Partial Differential Equations
• Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equations
• Physics-guided Neural Networks (PGNN): An Application in Lake Temperature Modeling
• PDE-Net: Learning PDEs from Data
• A neural network approach for the blind deconvolution of turbulent flows
• An Efficient Deep Learning Technique for the Navier-Stokes Equations: Application to Unsteady Wake Flow Dynamics
• Stability selection enables robust learning of partial differential equations from limited noisy data

Cited by in corpus (36)

• Physics-informed learning of governing equations from scarce data
• Discovering Symbolic Models from Deep Learning with Inductive Biases
• Unsupervised discovery of interpretable hyperelastic constitutive laws
• Biologically-informed neural networks guide mechanistic modeling from sparse experimental data
• Learning hydrodynamic equations for active matter from particle simulations and experiments
• Digital Twin: Values, Challenges and Enablers
• Physics informed deep learning for computational elastodynamics without labeled data
• Noise-aware Physics-informed Machine Learning for Robust PDE Discovery
• Physics-constrained Deep Learning for Robust Inverse ECG Modeling
• Augmenting astrophysical scaling relations with machine learning: application to reducing the Sunyaev-Zeldovich flux-mass scatter
• A Method for Computing Inverse Parametric PDE Problems with Random-Weight Neural Networks
• Discovery of partial differential equations from highly noisy and sparse data with physics-informed information criterion
• Learning physically consistent mathematical models from data using group sparsity
• Parsimony-Enhanced Sparse Bayesian Learning for Robust Discovery of Partial Differential Equations
• Hypersolvers: Toward Fast Continuous-Depth Models
• Discovering dynamic laws from observations: the case of self-propelled, interacting colloids
• Deep Euler method: solving ODEs by approximating the local truncation error of the Euler method
• Learning minimal representations of stochastic processes with variational autoencoders
• Temporal Normalizing Flows
• SINDy-RL: Interpretable and Efficient Model-Based Reinforcement Learning
• Sparsely constrained neural networks for model discovery of PDEs
• Weak Collocation Regression method: fast reveal hidden stochastic dynamics from high-dimensional aggregate data
• Learning Dynamics from Noisy Measurements using Deep Learning with a Runge-Kutta Constraint
• Learning to Discretize: Solving 1D Scalar Conservation Laws via Deep Reinforcement Learning
• Robust discovery of partial differential equations in complex situations
• Physics-informed AI and ML-based sparse system identification algorithm for discovery of PDE's representing nonlinear dynamic systems
• Electron neural closure for turbulent magnetosheath simulations: energy channels
• Physics-informed Spline Learning for Nonlinear Dynamics Discovery
• Data-Driven Theory-guided Learning of Partial Differential Equations using SimultaNeous Basis Function Approximation and Parameter Estimation (SNAPE)
• Sparsistent Model Discovery
• Identifying Physical Law of Hamiltonian Systems via Meta-Learning
• Model discovery in the sparse sampling regime
• Discovering PDEs from Multiple Experiments
• The Machine Learning Approach to Moment Closure Relations for Plasma: A Review
• Fully differentiable model discovery
• Quantum Model-Discovery