AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Learning Interface Conditions in Domain Decomposition Solvers

Ali Taghibakhshi, Nicolas Nytko, Tareq Uz Zaman, Scott MacLachlan, Luke Olson, Matthew West

NeurIPS 2022 · Learning theory and algorithms › Combinatorial and search optimization · approach: Constrained neural network optimization · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Learn interface conditions with MLORAS instead of using RASHelmholtz problem on unstructured triangular grids · Unstructured-grid test setFGMRES convergenceImproved
Learn interface conditions with MLORAS instead of using RASHelmholtz problem on unstructured triangular grids · Unstructured-grid test setstationary iteration convergenceImproved
Use MLORAS rather than RASHelmholtz problem on a 100×100 structured grid · 10 stationary iterationsL2 norm of error: 6.526 -> 0.231Improved
Use MLORAS rather than RAS as the preconditionerFGMRES for the Helmholtz problem on a 100×100 structured grid · 10 FGMRES stepsL2 norm of error: 0.146 -> 0.084Improved
Use the sampled power-iteration loss instead of Frobenius-norm minimizationMLORAS training · Unstructured-grid test setconvergence rateImproved
Use Frobenius-norm minimization instead of the sampled power-iteration loss (ablation)MLORAS training · Unstructured-grid test setFrobenius normImproved
Increase residual block length (ablation)MLORAS GNN with 4 TAGConv layers · 50 unstructured grids of 2400 to 2600 nodesFGMRES stepsImproved
Use more than 4 TAGConv layers (ablation)MLORAS GNN with residual block length 8 · 50 unstructured grids of 2400 to 2600 nodesFGMRES stepsHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy