AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Swarm Reinforcement Learning for Adaptive Mesh Refinement

Niklas Freymuth, Philipp Dahlinger, Tobias Daniel Würth, Simon Reisch, Luise Kärger, Gerhard Neumann

NeurIPS 2023 · Reinforcement learning › Visual and representation RL · approach: RL with tree search for discovery · 9 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
combine per-element spatial reward with area scaling (ablation)ASMR · Linear Elasticity (reward ablation)normalized squared error vs number of elementsImproved
add area scaling term to a global (non-spatial) reward (ablation)ASMR · Linear Elasticity (reward ablation)normalized squared error vs number of elementsHurt
replace mean-error reward with maximum-error reward (ablation)ASMR · Linear Elasticityrefinement qualityNo clear effect
use PPO instead of DQN (ablation)ASMR and learned baselines · Poissonmesh qualityImproved
add absolute positions to node features (ablation)ASMR · Poissonmesh qualityHurt
add solution information and load function evaluations to node features (ablation)ASMR · Poissonmesh qualityImproved
increase training PDEs beyond 100 (ablation)ASMR · Poissonmesh qualityNo clear effect
use MPN instead of GAT-like architecture (ablation)VDGN · Poissonmesh qualityImproved
use ASMR refinement instead of uniform meshPoisson (large spiral domains) · Poisson spiral domainswall-clock speedup: more than factor 100Improved
use ASMR refinement instead of uniform meshall tasks · all taskswall-clock speedup: factor 2 to 30Improved
use ASMR instead of oracle/ZZ heuristicsmesh refinement · Stokes flowerror vs number of elementsHurt
use ASMR instead of learned baselines at 6 refinement stepsmesh refinement · Laplace, 6 refinement stepserror vs number of elementsImproved

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