Scalable Data-Driven Basis Selection for Linear Machine Learning Interatomic Potentials

Abstract

Machine learning interatomic potentials (MLIPs) provide an effective approach for accurately and efficiently modeling atomic interactions, expanding the capabilities of atomistic simulations to complex systems. However, a priori feature selection leads to high complexity, which can be detrimental to both computational cost and generalization, resulting in a need for hyperparameter tuning. We demonstrate the benefits of active set algorithms for automated data-driven feature selection. The proposed methods are implemented within the Atomic Cluster Expansion (ACE) framework. Computational tests conducted on a variety of benchmark datasets indicate that sparse ACE models consistently enhance computational efficiency, generalization accuracy and interpretability over dense ACE models. An added benefit of the proposed algorithms is that they produce entire paths of models with varying cost/accuracy ratio.

References in corpus (28)

• Least Angle Regression
• Rejoinder to "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• On representing chemical environments
• Perspective on density functional theory
• Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals
• Moment Tensor Potentials: a class of systematically improvable interatomic potentials
• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• A Performance and Cost Assessment of Machine Learning Interatomic Potentials
• On asymptotically optimal confidence regions and tests for high-dimensional models
• Machine learning a general purpose interatomic potential for silicon
• Ab initio thermodynamics of liquid and solid water
• De novo exploration and self-guided learning of potential-energy surfaces
• Learning Properties of Ordered and Disordered Materials from Multi-fidelity Data
• A sparse representation for potential energy surface
• First-principles interatomic potentials for ten elemental metals via compressed sensing
• Tensor-reduced atomic density representations
• Enhancing the Accuracy of Density Functional Tight Binding Models Through ChIMES Many-body Interaction Potentials
• Measuring transferability issues in machine-learning force fields: The example of Gold-Iron interactions with linearized potentials