Skip to Main content Skip to Navigation
Conference papers

Solver-in-the-Loop: Learning from Differentiable Physics to Interact with Iterative PDE-Solvers

Abstract : Finding accurate solutions to partial differential equations (PDEs) is a crucial task in all scientific and engineering disciplines. It has recently been shown that machine learning methods can improve the solution accuracy by correcting for effects not captured by the discretized PDE. We target the problem of reducing numerical errors of iterative PDE solvers and compare different learning approaches for finding complex correction functions. We find that previously used learning approaches are significantly outperformed by methods that integrate the solver into the training loop and thereby allow the model to interact with the PDE during training. This provides the model with realistic input distributions that take previous corrections into account, yielding improvements in accuracy with stable rollouts of several hundred recurrent evaluation steps and surpassing even tailored supervised variants. We highlight the performance of the differentiable physics networks for a wide variety of PDEs, from non-linear advection-diffusion systems to three-dimensional Navier-Stokes flows.
Complete list of metadata

https://hal.telecom-paris.fr/hal-03142245
Contributor : Kiwon Um <>
Submitted on : Monday, February 15, 2021 - 8:59:26 PM
Last modification on : Tuesday, September 21, 2021 - 2:16:04 PM

Links full text

Identifiers

  • HAL Id : hal-03142245, version 1
  • ARXIV : 2007.00016

Collections

Citation

Kiwon Um, Robert Brand, Fei Yun, Philipp Holl, Nils Thuerey. Solver-in-the-Loop: Learning from Differentiable Physics to Interact with Iterative PDE-Solvers. Neural Information Processing Systems (NeurIPS), Dec 2020, Vancouver, Canada. pp.6111-6122. ⟨hal-03142245⟩

Share

Metrics

Record views

66