# Learning to Optimize Multigrid PDE Solvers

@inproceedings{Greenfeld2019LearningTO, title={Learning to Optimize Multigrid PDE Solvers}, author={Daniel Greenfeld and Meirav Galun and Ronen Basri and Irad Yavneh and Ron Kimmel}, booktitle={ICML}, year={2019} }

Constructing fast numerical solvers for partial differential equations (PDEs) is crucial for many scientific disciplines. [...] Key Method Our method learns a (single) mapping from discretized PDEs to prolongation operators for a broad class of 2D diffusion problems. We train a neural network once for the entire class of PDEs, using an efficient and unsupervised loss function. Our tests demonstrate improved convergence rates compared to the widely used Black-Box multigrid scheme, suggesting that our method…Expand Abstract

Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

2

Twitter Mentions

#### Citations

##### Publications citing this paper.

SHOWING 1-4 OF 4 CITATIONS

## Learning Algebraic Multigrid Using Graph Neural Networks

VIEW 7 EXCERPTS

CITES BACKGROUND & METHODS

#### References

##### Publications referenced by this paper.

SHOWING 1-10 OF 23 REFERENCES

## A machine learning framework for data driven acceleration of computations of differential equations

VIEW 12 EXCERPTS

HIGHLY INFLUENTIAL

## General solutions for nonlinear differential equations: a deep reinforcement learning approach

VIEW 12 EXCERPTS

HIGHLY INFLUENTIAL

## Overcoming the curse of dimensionality: Solving high-dimensional partial differential equations using deep learning

VIEW 11 EXCERPTS

HIGHLY INFLUENTIAL

## Study on a Poisson's equation solver based on deep learning technique

VIEW 12 EXCERPTS

HIGHLY INFLUENTIAL

## Artificial Neural Networks Approach for Solving Stokes Problem

VIEW 11 EXCERPTS

HIGHLY INFLUENTIAL

## Learning Neural PDE Solvers with Convergence Guarantees

VIEW 3 EXCERPTS

## Fourier Analysis of Periodic Stencils in Multigrid Methods

VIEW 2 EXCERPTS