Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Message Passing Neural PDE Solvers

About

The numerical solution of partial differential equations (PDEs) is difficult, having led to a century of research so far. Recently, there have been pushes to build neural--numerical hybrid solvers, which piggy-backs the modern trend towards fully end-to-end learned systems. Most works so far can only generalize over a subset of properties to which a generic solver would be faced, including: resolution, topology, geometry, boundary conditions, domain discretization regularity, dimensionality, etc. In this work, we build a solver, satisfying these properties, where all the components are based on neural message passing, replacing all heuristically designed components in the computation graph with backprop-optimized neural function approximators. We show that neural message passing solvers representationally contain some classical methods, such as finite differences, finite volumes, and WENO schemes. In order to encourage stability in training autoregressive models, we put forward a method that is based on the principle of zero-stability, posing stability as a domain adaptation problem. We validate our method on various fluid-like flow problems, demonstrating fast, stable, and accurate performance across different domain topologies, equation parameters, discretizations, etc., in 1D and 2D.

Johannes Brandstetter, Daniel Worrall, Max Welling• 2022

Related benchmarks

TaskDatasetResultRank
Zero-shot super-resolutionE1 (test)
MAE0.1541
48
Learning PDE Solution Operators2D Shallow Water
Mean L2 Relative Error0.5
20
Surrogate modeling of 1D nonlinear PDEs1D family of nonlinear PDEs Scenario E1 1.0 (test)
Accumulated Error (MSE)0.51
15
Surrogate modeling of 1D nonlinear PDEs1D family of nonlinear PDEs Scenario E3 1.0 (test)
Accumulated Error (MSE)3.7
15
Surrogate modeling of 1D nonlinear PDEs1D family of nonlinear PDEs Scenario E2 1.0 (test)
Accumulated Error (MSE)1.45
15
Learning PDE Solution Operators1D Diffusion-Reaction
Mean L2 Rel Error39
12
Learning PDE Solution OperatorsAllen-Cahn 1D
Mean L2 Relative Error0.33
12
Learning PDE Solution Operators1D Cahn-Hilliard
Mean L2 Relative Error0.59
8
Learning PDE Solution Operators1D Diffusion-Sorption
Mean L2 Relative Error2.86
8
Learning PDE Solution Operators3D Maxwell
Mean L2 Relative Error0.7445
8
Showing 10 of 10 rows

Other info

Follow for update