Compact atomic descriptors enable accurate predictions via linear models

Abstract

We probe the accuracy of linear ridge regression employing a three-body local density representation derived from the atomic cluster expansion. We benchmark the accuracy of this framework in the prediction of formation energies and atomic forces in molecules and solids. We find that such a simple regression framework performs on par with state-of-the-art machine learning methods which are, in most cases, more complex and more computationally demanding. Subsequently, we look for ways to sparsify the descriptor and further improve the computational efficiency of the method. To this aim, we use both principal component analysis and least absolute shrinkage operator regression for energy fitting on six single-element datasets. Both methods highlight the possibility of constructing a descriptor that is four times smaller than the original with a similar or even improved accuracy. Furthermore, we find that the reduced descriptors share a sizable fraction of their features across the six independent datasets, hinting at the possibility of designing material-agnostic, optimally compressed, and accurate descriptors.

10 pages, 8 figures

References in corpus (8)

• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• Big Data of Materials Science - Critical Role of the Descriptor
• Efficient and Accurate Machine-Learning Interpolation of Atomic Energies in Compositions with Many Species
• Compressive sensing as a new paradigm for model building
• Ranking the information content of distance measures
• A Novel Approach to Describe Chemical Environments in High Dimensional Neural Network Potentials
• Hydration free energies from kernel-based machine learning: Compound-database bias

Cited by in corpus (7)

• Data-driven simulation and characterisation of gold nanoparticle melting
• Tensor-reduced atomic density representations
• Exploring the robust extrapolation of high-dimensional machine learning potentials
• A Physically-informed Graph-based Order Parameter for the Universal Characterization of Atomic Structures
• Through the eyes of a descriptor: Constructing complete, invertible descriptions of atomic environments
• Automatic feature selection and weighting in molecular systems using Differentiable Information Imbalance
• Accelerating the theoretical study of Li-polysulphide adsorption on single-atom catalysts via machine learning approaches