Facilitating {\it ab initio} configurational sampling of multicomponent solids using an on-lattice neural network model and active learning
arXiv:2008.02572 · doi:10.1063/5.0096645
Abstract
We propose a scheme for {\it ab initio} configurational sampling in multicomponent crystalline solids using Behler-Parinello type neural network potentials (NNPs) in an unconventional way: the NNPs are trained to predict the energies of relaxed structures from the perfect lattice with configurational disorder instead of the usual way of training to predict energies as functions of continuous atom coordinates. An active learning scheme is employed to obtain a training set containing configurations of thermodynamic relevance. This enables bypassing of the structural relaxation procedure which is necessary when applying conventional NNP approaches to the lattice configuration problem. The idea is demonstrated on the calculation of the temperature dependence of the degree of A/B site inversion in three spinel oxides, MgAlO, ZnAlO, and MgGaO. The present scheme may serve as an alternative to cluster expansion for `difficult' systems, e.g., complex bulk or interface systems with many components and sublattices that are relevant to many technological applications today.
arXiv comment12 pages, 11 figures, revised title, added discussion and benchmark examples
References in corpus (30)
- Global Optimization by Basin-Hopping and the Lowest Energy Structures of Lennard-Jones Clusters Containing up to 110 Atoms
- An efficient, multiple range random walk algorithm to calculate the density of states
- Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
- On representing chemical environments
- Exchange Monte Carlo Method and Application to Spin Glass Simulations
- Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics
- SchNet - a deep learning architecture for molecules and materials
- The Alloy Theoretic Automated Toolkit: A User Guide
- Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals
- Verification of Short-Range Order and Its Impact on the Properties of the CrCoNi Medium Entropy Alloy
- A Performance and Cost Assessment of Machine Learning Interatomic Potentials
- Less is more: sampling chemical space with active learning
- The Open Catalyst 2020 (OC20) Dataset and Community Challenges
- Active learning of linearly parametrized interatomic potentials
- Active Learning of Uniformly Accurate Inter-atomic Potentials for Materials Simulation
- Towards Universal Neural Network Potential for Material Discovery Applicable to Arbitrary Combination of 45 Elements
- Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning
- Efficient and Accurate Machine-Learning Interpolation of Atomic Energies in Compositions with Many Species
- Impact of lattice relaxations on phase transitions in a high-entropy alloy studied by machine-learning potentials
- End-to-end Symmetry Preserving Inter-atomic Potential Energy Model for Finite and Extended Systems
- TeaNet: universal neural network interatomic potential inspired by iterative electronic relaxations
- Accurate representation of formation energies of crystalline alloys with many components
- CLEASE: A versatile and user-friendly implementation of Cluster Expansion method
- The Robustness of Cluster Expansion: Assessing the Roles of Relaxation and Numerical Error
- Density-functional Monte-Carlo simulation of CuZn order-disorder transition
- Self-learning Hybrid Monte Carlo: A First-principles Approach
- Robust cluster expansion of multicomponent systems using structured sparsity
- Self-learning Monte Carlo method with Behler-Parrinello neural networks
- An Introduction to Electrocatalyst Design using Machine Learning for Renewable Energy Storage
- Direct coupling of first-principles calculations with replica exchange Monte Carlo sampling of ion disorder in solids