mlsanityData to 14 Oct 2026

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

arXiv 2211.12020 ↗First posted 22 Nov 2022cs.LG physics.comp-phcited by 6

Not on the map. Why?

Abstract (arXiv)

Mitigating the climate crisis requires a rapid transition towards lower-carbon energy. Catalyst materials play a crucial role in the electrochemical reactions involved in numerous industrial processes key to this transition, such as renewable energy storage and electrofuel synthesis. To reduce the energy spent on such activities, we must quickly discover more efficient catalysts to drive electrochemical reactions. Machine learning (ML) holds the potential to efficiently model materials properties from large amounts of data, accelerating electrocatalyst design. The Open Catalyst Project OC20 dataset was constructed to that end. However, ML models trained on OC20 are still neither scalable nor accurate enough for practical applications. In this paper, we propose task-specific innovations applicable to most architectures, enhancing both computational efficiency and accuracy. This includes improvements in (1) the graph creation step, (2) atom representations, (3) the energy prediction head, and (4) the force prediction head. We describe these contributions, referred to as PhAST, and evaluate them thoroughly on multiple architectures. Overall, PhAST improves energy MAE by 4 to 42% while dividing compute time by 3 to 8× depending on the targeted task/model. PhAST also enables CPU training, leading to 40× speedups in highly parallelized settings. Python package: urlhttps://phast.readthedocs.io.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 24 of its 41 references that are in the map. First posted on arXiv 22 Nov 2022.

Builds on

24 of 41 references.

References · 24

  1. 2017Attention Is All You Needcited by 60,361
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  3. 2021E(n) Equivariant Graph Neural Networkscited by 965
  4. 2015Convolutional Networks on Graphs for Learning Molecular Fingerprintscited by 754
  5. 2018Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point cloudscited by 746
  6. 2021E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentialscited by 708
  7. 2020SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networkscited by 607
  8. 2017SchNet: A continuous-filter convolutional neural network for modeling quantum interactionscited by 571
  9. 2020Directional Message Passing for Molecular Graphscited by 561
  10. 2021Equivariant message passing for the prediction of tensorial properties and molecular spectracited by 441
  11. 2021GemNet: Universal Directional Graph Neural Networks for Moleculescited by 343
  12. 2019Cormorant: Covariant Molecular Neural Networkscited by 263
12 more
  1. 2020Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Moleculescited by 255
  2. 2020The Open Catalyst 2020 (OC20) Dataset and Community Challengescited by 246
  3. 2021Geometric and Physical Quantities Improve E(3) Equivariant Message Passingcited by 228
  4. 2022TorchMD-NET: Equivariant Transformers for Neural Network based Molecular Potentialscited by 182
  5. 2022GemNet-OC: Developing Graph Neural Networks for Large and Diverse Molecular Simulation Datasetscited by 61
  6. 2021ForceNet: A Graph Neural Network for Large-Scale Quantum Calculationscited by 58
  7. 2021Rotation Invariant Graph Neural Networks using Spin Convolutionscited by 52
  8. 2022Spherical Channels for Modeling Atomic Interactionscited by 47
  9. 2020An Introduction to Electrocatalyst Design using Machine Learning for Renewable Energy Storagecited by 45
  10. 2021Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotationscited by 29
  11. 2022Higher-order Clustering and Pooling for Graph Neural Networkscited by 26
  12. 2022The Open MatSci ML Toolkit: A Flexible Framework for Machine Learning in Materials Sciencecited by 14

Built on by

Cited by year

Citing papers per year: 2023: 4; 2024: 1; 2025: 1.

6 papers.

  1. 2023FAENet: Frame Averaging Equivariant GNN for Materials Modelingcited by 73
  2. 2023A Hitchhiker's Guide to Geometric GNNs for 3D Atomic Systemscited by 68
  3. 2023Crystal-GFN: sampling crystals with desirable properties and constraintscited by 31
  4. 2023On the importance of catalyst-adsorbate 3D interactions for relaxed energy predictionscited by 1
  5. 2024Improving Molecular Modeling with Geometric GNNs: an Empirical Studycited by 0
  6. 2025Differentiated Information Mining: A Semi-supervised Learning Framework for GNNscited by 0