Learning Smooth and Expressive Interatomic Potentials for Physical Property Prediction
Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:EI7HBS7Grecord.jsonopen to challenge →
read the original abstract
Machine learning interatomic potentials (MLIPs) have become increasingly effective at approximating quantum mechanical calculations at a fraction of the computational cost. However, lower errors on held out test sets do not always translate to improved results on downstream physical property prediction tasks. In this paper, we propose testing MLIPs on their practical ability to conserve energy during molecular dynamic simulations. If passed, improved correlations are found between test errors and their performance on physical property prediction tasks. We identify choices which may lead to models failing this test, and use these observations to improve upon highly-expressive models. The resulting model, eSEN, provides state-of-the-art results on a range of physical property prediction tasks, including materials stability prediction, thermal conductivity prediction, and phonon calculations.
This paper has not been read by Pith yet.
Forward citations
Cited by 15 Pith papers
-
Surrogate-Gated Generation and Foundation-Model Embeddings for Bayesian Materials Design
A Gaussian process surrogate gate inserted between generative crystal models and property oracles matches or exceeds ungated fine-tuning while using roughly one-fifth the oracle calls for heat capacity and bulk modulus.
-
Unravelling the Role of Stacking Disorder on the Optoelectronic Properties of Zn3P2
Planar faults in Zn3P2 arise from flexible vacant-site ordering, exhibit 2.5 mJ m^{-2} formation energy, introduce no mid-gap states, and may act as segregation sites for optically active point defects.
-
Fast contracted Clebsch--Gordan tensor products for equivariant graph neural networks
An O(L^3) algorithm computes contracted Clebsch-Gordan tensor products for equivariant ML potentials using a structured angular grid and spherical Poisson bracket to handle parity-odd terms at fixed CP rank.
-
OmniMol: Transferring Particle Physics Knowledge to Molecular Dynamics with Point-Edge Transformers
OmniMol transfers a billion-jet pre-trained PET foundation model from HEP to molecular dynamics via an interaction-matrix attention bias, delivering strong performance on the oMol dataset with minimal fine-tuning and ...
-
Approaching the Limit of Intrinsic Crystalline Thermal Insulation
High-throughput ML interatomic potentials plus phonon calculations discover and experimentally confirm CsTlI4 as having record-low intrinsic room-temperature thermal conductivity of 0.14 W m^{-1} K^{-1}.
-
Tensor Channel Equivariant Graph Neural Networks for Molecular Polarizability Prediction
A tensor-channel equivariant GNN based on PaiNN propagates symmetric rank-2 tensor features during message passing and achieves lower full-tensor and anisotropic error than readout-only and MACE baselines on QM7-X geometries.
-
Categorification of Chemical Reactions: a bottom-up tower from stoichiometry to quantum structure
Builds a canonical tower of nine categorical levels for chemical reactions from stoichiometry to quantum mechanics, with each level uniquely determined by cokernels in automorphism sequences and recovering Feinberg's ...
-
Crys-JEPA: Accelerating Crystal Discovery via Embedding Screening and Generative Refinement
Crys-JEPA introduces a joint embedding predictive architecture that creates an energy-aware latent space, enabling embedding-based stability screening and a refinement pipeline that yields up to 72.7% gains on the V.S...
-
Selectivity- and Activity-Aware Catalyst Descriptors for CO$_2$ Hydrogenation on Alloy Nanocatalysts using Machine-Learned Force Fields
A facet-resolved adsorption energy distribution method with ML force fields identifies active and methanol-selective alloy nanocatalyst surfaces for CO2 hydrogenation.
-
VibroML: an automated toolkit for high-throughput vibrational analysis and dynamic instability remediation of crystalline materials using machine-learned potentials
VibroML automates remediation of dynamic instabilities in crystalline materials by combining MLIPs with genetic algorithms for polymorph search, finite-temperature MD validation, and compositional alloying to yield st...
-
UBio-MolFM: A Universal Molecular Foundation Model for Bio-Systems
UBio-MolFM achieves ab initio-level fidelity on large out-of-distribution biomolecular systems using a new multi-fidelity dataset, E2Former-V2 architecture, and three-stage curriculum learning.
-
AI-Driven Expansion and Application of the Alexandria Database
A combined generative model, ML potential, and graph neural network pipeline expands the Alexandria database by 1.3 million DFT-validated compounds with 99% success near the convex hull and releases training data for ...
-
Benchmarking empirical and machine-learned interatomic potentials using phase diagram predictions for Lead
The EDDP machine-learned potential for lead predicts the observed FCC-HCP phase transition at ~15 GPa, unlike EAM and MEAM models, when paired with nested sampling.
-
Machine Learning Interatomic Potentials: Advancing Open-Source Software for Efficient and Scalable Molecular Simulation
mlip v2 is a new software release that integrates API redesign, e3j backend, eSEN model, improved charge modeling, and expanded simulation capabilities to support larger-scale molecular modeling.
-
Six Open Questions in Machine-Learned Interatomic Potential Foundation Models
This perspective article develops a definition of foundational MLIPs and poses six open questions that the authors believe will define future research in machine-learned interatomic potentials.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.