Black-box learning of multigrid parameters

DOI10.1016/j.cam.2019.112524zbMath1440.65265OpenAlexW2980085231WikidataQ127114086 ScholiaQ127114086MaRDI QIDQ2291999

Alexandr Katrutsa, T. Daulbaev, Ivan V. Oseledets

Publication date: 31 January 2020

Published in: Journal of Computational and Applied Mathematics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.cam.2019.112524

zbMATH Keywords

automatic differentiationHelmholtz equationPoisson equationgeometric multigrid methodspectral radius minimization

Mathematics Subject Classification ID

Multigrid methods; domain decomposition for boundary value problems involving PDEs (65N55) Numerical optimization and variational techniques (65K10) Learning and adaptive systems in artificial intelligence (68T05) Stochastic approximation (62L20) Laplace operator, Helmholtz equation (reduced wave equation), Poisson equation (35J05) Finite difference methods for boundary value problems involving PDEs (65N06)

Related Items (8)

Spectral coarse spaces for the substructured parallel Schwarz method ⋮ Efficient hybrid explicit-implicit learning for multiscale problems ⋮ Learning-based local weighted least squares for algebraic multigrid method ⋮ Accelerating algebraic multigrid methods via artificial neural networks ⋮ Connections between numerical algorithms for PDEs and neural networks ⋮ Tuning Multigrid Methods with Robust Optimization and Local Fourier Analysis ⋮ Direct optimization of BPX preconditioners ⋮ Data-driven multi-grid solver for accelerated pressure projection

Uses Software

• Theano
• PyAMG
• TensorFlow
• DiffSharp
• PyTorch
• Autograd
• GitHub

Cites Work

• Unnamed Item
• Unnamed Item
• Fast and high accuracy multiscale multigrid method with multiple coarse grid updating strategy for the 3D convection-diffusion equation
• A nearly optimal multigrid method for general unstructured grids
• On accuracy of approximation of the spectral radius by the Gelfand formula
• Smoothing technique and its applications in semidefinite optimization
• Modern homotopy methods in optimization
• Black box multigrid
• The complexity of partial derivatives
• The auxiliary space method and optimal multigrid preconditioning techniques for unstructured grids
• A hybrid multigrid method for convection-diffusion problems
• On the order of prolongations and restrictions in multigrid procedures
• Matrix-dependent prolongations and restrictions in a blackbox multigrid solver
• Optimizing the Spectral Radius
• Subspace Methods for Computing the Pseudospectral Abscissa and the Stability Radius
• Numerical Optimization of Eigenvalues of Hermitian Matrix Functions
• Iterative Methods for Linear Systems
• Randomized algorithms for estimating the trace of an implicit symmetric positive semi-definite matrix
• Bootstrap AMG
• A General Interpolation Strategy for Algebraic Multigrid Using Energy Minimization
• A Multigrid Tutorial, Second Edition
• Adaptive Smoothed Aggregation ($\alpha$SA)
• A Genetic Search for Optimal Multigrid Components Within a Fourier Analysis Setting
• A Multigrid Method for the Helmholtz Equation with Optimized Coarse Grid Corrections
• Adaptive Algebraic Multigrid
• A scalable multigrid method for solving indefinite Helmholtz equations with constant wave numbers
• A stochastic estimator of the trace of the influence matrix for laplacian smoothing splines
• A review of algebraic multigrid

This page was built for publication: Black-box learning of multigrid parameters