AI Paper Atlas

Research map / Graph learning / GNN architectures

PhAST: Physics-Aware, Scalable, and Task-Specific GNNs for Accelerated Catalyst Design

Alexandre Duval, Victor Schmidt, Santiago Miret, Yoshua Bengio, Alex Hernández-García, David Rolnick

JMLR 2024 · Graph learning › GNN architectures · approach: Physics-inspired GNNs · 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
add all PhAST components (graph pruning, rich embeddings, weighted-sum energy head)SchNet · OC20 IS2REaverage energy MAE: 683 -> 630Improved
add all PhAST componentsSchNet · OC20 IS2REinference time (s): 15 -> 5Improved
add all PhAST componentsGemNet-OC · OC20 IS2REinference time (s): 868 -> 257Improved
add tag embeddings to atom embeddings (ablation)SchNet · OC20 IS2REaverage energy MAE: 683 -> 648Improved
remove tag-0 sub-surface atoms from the graph (ablation)SchNet · OC20 IS2REinference time (s): 15.40 -> 4.74Improved
remove tag-0 sub-surface atoms from the graph (ablation)DimeNet++ · OC20 IS2REinference time (s): 110.11 -> 20.15Improved
remove tag-0 sub-surface atoms from the graph (ablation)DimeNet++ · OC20 IS2REaverage energy MAE: 628 -> 610Improved
add one supernode per graph instead of removing tag-0 (ablation)SchNet · OC20 IS2REaverage energy MAE: 648 -> 654No clear effect
add hierarchical HoscPool energy head (ablation)SchNet · OC20 IS2REinference time (s): 15.40 -> 53.93Hurt
add weighted-sum energy head from final embeddings (ablation)ForceNet · OC20 IS2REaverage energy MAE: 654 -> 660Hurt
add weighted-sum energy head from initial embeddings (ablation)SchNet · OC20 IS2REaverage energy MAE: 683 -> 657Improved
replace gradient-based forces with direct force head (PhAST-Direct) (ablation)SchNet · OC20 S2EF-2Menergy MAE vs force MAE: E-MAE 890 -> 716; F-MAE 68.4 -> 81.2Mixed

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