AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

Guaranteed Conservation of Momentum for Learning Particle-based Fluid Dynamics

Lukas Prantl, Benjamin Ummenhofer, Vladlen Koltun, Nils Thuerey

NeurIPS 2022 · Probabilistic and scientific modeling › Dynamical systems · approach: Physics-informed dynamics learning · 12 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
add the antisymmetric output layer (ablation)Base network · WBC-SPHvelocity-distribution accuracyImproved
add the antisymmetric output layer (ablation)Base network · WBC-SPHchange in momentum: 0.0 -> 1.0Improved
use multi-scale processing with farthest-point sampling (ablation)ASCC · WBC-SPHoverall performanceMixed
replace farthest-point sampling with voxelization (ablation)Multi Scale · WBC-SPHresource usageImproved
add temporal preprocessing rollouts (ablation)Voxelize · WBC-SPHaverage relative performance: 0.94 -> 1.0Improved
add input rotation normalization and enhanced boundary processing (ablation)Preprocess · WBC-SPHaverage relative performance: 0.94 -> 1.0Improved
replace ASCC with regular CConv (ablation)Ours · Liquid column, Free FallRMSE: 19.68 -> 46.36843Hurt
train with a 20-step rather than 5-step rollout (ablation)Ours · WaterRampsEMD: 0.09155 -> 0.06156Improved
use the proposed method instead of GNSWaterRamps · WaterRampsEMD: 24%Improved
use the proposed method instead of GNSOurs · Droplet collisions, default gravityEMD: 57%Improved
change gravity to zero at evaluationGNS · Droplet collisions, zero gravityEMD: 195% larger error than our methodHurt
train with the momentum-conserving constraint (ablation)ASCC · WBC-SPH sample-efficiency evaluationEMD: ca. 6% ... more than 20%Improved

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