Gradient-based grand canonical optimization enabled by graph neural networks with fractional atomic existence
Open Access
- 18 November 2025
- journal article
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 6 (4) , 045049
Abstract
Machine learning interatomic potentials have become an indispensable tool for materials science, enabling the study of larger systems and longer timescales. State-of-the-art models are generally graph neural networks that employ message passing to iteratively update atomic embeddings that are ultimately used for predicting properties. In this work we extend the message passing formalism with the inclusion of a continuous variable that accounts for fractional atomic existence. This allows us to calculate the gradient of the Gibbs free energy with respect to both the Cartesian coordinates of atoms and their existence. Using this we propose a gradient-based grand canonical optimization method and document its capabilities for a Cu(110) surface oxide.Keywords
All Related Versions
Funding Information
- Danmarks Grundforskningsfond (150)
- Villum Fonden (16562)
This publication has 65 references indexed in Scilit:
- Scalable Parallel Algorithm for Graph Neural Network Interatomic Potentials in Molecular Dynamics SimulationsJournal of Chemical Theory and Computation, 2024
- Clinamen2: Functional-style evolutionary optimization in Python for atomistic structure searchesComputer Physics Communications, 2023
- Exploring Potential Energy Surfaces Using Reinforcement Machine LearningJournal of Chemical Information and Modeling, 2022
- The XtalOpt Evolutionary Algorithm for Crystal Structure PredictionThe Journal of Physical Chemistry C, 2020
- PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments, and Partial ChargesJournal of Chemical Theory and Computation, 2019
- CALYPSO structure prediction method and its wide applicationComputational Materials Science, 2016
- A Comparison of Advanced Monte Carlo Methods for Open Systems: CFCMC vs CBMCJournal of Chemical Theory and Computation, 2014
- Genetic Algorithm Procreation Operators for Alloy Nanoparticle CatalystsTopics in Catalysis, 2013
- Continuous Fractional Component Monte Carlo: An Adaptive Biasing Method for Open System Atomistic SimulationsJournal of Chemical Theory and Computation, 2007
- Global minimum structure searches via particle swarm optimizationJournal of Computational Chemistry, 2007