Monte Carlo Neural PDE Solver for Learning PDEs via Probabilistic Representation

Abstract

In scenarios with limited available data, training the function-to-function neural PDE solver in an unsupervised manner is essential. However, the efficiency and accuracy of existing methods are constrained by the properties of numerical algorithms, such as finite difference and pseudo-spectral methods, integrated during the training stage. These methods necessitate careful spatiotemporal discretization to achieve reasonable accuracy, leading to significant computational challenges and inaccurate simulations, particularly in cases with substantial spatiotemporal variations. To address these limitations, we propose the Monte Carlo Neural PDE Solver (MCNP Solver) for training unsupervised neural solvers via the PDEs' probabilistic representation, which regards macroscopic phenomena as ensembles of random particles. Compared to other unsupervised methods, MCNP Solver naturally inherits the advantages of the Monte Carlo method, which is robust against spatiotemporal variations and can tolerate coarse step size. In simulating the trajectories of particles, we employ Heun's method for the convection process and calculate the expectation via the probability density function of neighbouring grid points during the diffusion process. These techniques enhance accuracy and circumvent the computational issues associated with Monte Carlo sampling. Our numerical experiments on convection-diffusion, Allen-Cahn, and Navier-Stokes equations demonstrate significant improvements in accuracy and efficiency compared to other unsupervised baselines. The source code will be publicly available at: https://github.com/optray/MCNP.

IEEE TPAMI 2025

References in corpus (22)

• DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators
• An Energy Approach to the Solution of Partial Differential Equations in Computational Mechanics via Machine Learning: Concepts, Implementation and Applications
• Solving high-dimensional partial differential equations using deep learning
• NSFnets (Navier-Stokes Flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations
• Self-adaptive loss balanced Physics-informed neural networks for the incompressible Navier-Stokes equations
• Deep neural network solution of the electronic Schrödinger equation
• A Physics Informed Neural Network for Time-Dependent Nonlinear and Higher Order Partial Differential Equations
• Multi-scale Deep Neural Network (MscaleDNN) for Solving Poisson-Boltzmann Equation in Complex Domains
• Hybrid FEM-NN models: Combining artificial neural networks with the finite element method
• CENN: Conservative energy method based on neural networks with subdomains for solving variational problems involving heterogeneous and complex geometries
• Integrated Finite Element Neural Network (I-FENN) for non-local continuum damage mechanics
• Brownian-Time Processes: The PDE Connection and the Half-Derivative Generator
• A Derivative-Free Method for Solving Elliptic Partial Differential Equations with Deep Neural Networks
• Teaching the Incompressible Navier-Stokes Equations to Fast Neural Surrogate Models in 3D
• Deep Learning Methods for Partial Differential Equations and Related Parameter Identification Problems
• SVD Perspectives for Augmenting DeepONet Flexibility and Interpretability
• A neural network multigrid solver for the Navier-Stokes equations
• Stochastic Lagrangian Particle Approach to Fractal Navier-Stokes Equations
• Stochastic functional differential equations driven by Lévy processes and quasi-linear partial integro-differential equations
• Stochastic projection based approach for gradient free physics informed learning
• Deciphering and integrating invariants for neural operator learning with various physical mechanisms
• Exponentially slow motion of interface layers for the one-dimensional Allen-Cahn equation with nonlinear phase-dependent diffusivity