Lune

ICML2024Top-tier venue

Predicting and Interpreting Energy Barriers of Metallic Glasses with Graph Neural Networks

Haoyu Li, Shichang Zhang, Longwen Tang, Mathieu Bauchy, Yizhou Sun

Abstract

Metallic Glasses (MGs) are widely used materials that are stronger than steel while being shapeable as plastic. While understanding the structure-property relationship of MGs remains a challenge in materials science, studying their energy barriers (EBs) as an intermediary step shows promise. In this work, we utilize Graph Neural Networks (GNNs) to model MGs and study EBs. We contribute a new dataset for EB prediction and a novel Symmetrized GNN (SymGNN) model that is E(3)-invariant in expectation. SymGNN handles invariance by aggregating over orthogonal transformations of the graph structure. When applied to EB prediction, SymGNN are more accurate than molecular dynamics (MD) local-sampling methods and other machine-learning models. Compared to precise MD simulations, SymGNN reduces the inference time on new MGs from roughly 41 days to less than one second. We apply explanation algorithms to reveal the relationship between structures and EBs. The structures that we identify through explanations match the medium-range order (MRO) hypothesis and possess unique topological properties. Our work enables effective prediction and interpretation of MG EBs, bolstering material science research.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs cites Directional Message Passing for Molecular Graphs
  • Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs cites E(n) Equivariant Graph Neural Networks
  • Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs cites Equivariant message passing for the prediction of tensorial properties and molecular spectra
  • FAENet: Frame Averaging Equivariant GNN for Materials Modeling cites Directional Message Passing for Molecular Graphs
  • FAENet: Frame Averaging Equivariant GNN for Materials Modeling cites E(n) Equivariant Graph Neural Networks
  • FAENet: Frame Averaging Equivariant GNN for Materials Modeling cites Equivariant message passing for the prediction of tensorial properties and molecular spectra
  • E(n) Equivariant Graph Neural Networks cites Directional Message Passing for Molecular Graphs
  • Equivariant message passing for the prediction of tensorial properties and molecular spectra cites Directional Message Passing for Molecular Graphs
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on5

  1. E(n) Equivariant Graph Neural NetworksVictor Garcia Satorras, Emiel Hoogeboom, Max WellingICML 2021 · 1,432 citations
  2. Directional Message Passing for Molecular GraphsJohannes Klicpera, Janek Groß, Stephan GünnemannICLR 2020 · 1,079 citations
  3. Equivariant message passing for the prediction of tensorial properties and molecular spectraKristof Schütt, Oliver T. Unke, Michael GasteggerICML 2021 · 736 citations
  4. FAENet: Frame Averaging Equivariant GNN for Materials ModelingAlexandre Duval, Victor Schmidt, Alex Hernández-García, Santiago Miret et al.ICML 2023 · 93 citations
  5. Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic GraphsYi-Lun Liao, Tess E. SmidtICLR 2023 · 65 citations

Related papers

  1. Equivariant Networks for Crystal StructuresSékou-Oumar Kaba, Siamak RavanbakhshNeurIPS 2022 · 38 citations
  2. Conformal Prediction Sets for Graph Neural NetworksSoroush H. Zargarbashi, Simone Antonelli, Aleksandar BojchevskiICML 2023 · 49 citations
  3. Geometric Graph Neural Diffusion for Stable Molecular Dynamics SimulationsHaokai Hong, Wanyu Lin, Zhang Chusong, KC TanICLR 2026
  4. Equivariant Graph Mechanics Networks with ConstraintsWenbing Huang, Jiaqi Han, Yu Rong, Tingyang Xu et al.ICLR 2022 · 107 citations
  5. Learning the Electronic Hamiltonian of Large Atomic StructuresChen Hao Xia, Manasa Kaniselvan, Alexandros Nikolaos Ziogas, Marko Mladenovic et al.ICML 2025
Dusk over the sea between two cliffs drawn in fine vertical lines