Lightweight and Effective Tensor Sensitivity for Atomistic Neural Networks

Abstract

Atomistic machine learning focuses on the creation of models which obey fundamental symmetries of atomistic configurations, such as permutation, translation, and rotation invariances. In many of these schemes, translation and rotation invariance are achieved by building on scalar invariants, e.g., distances between atom pairs. There is growing interest in molecular representations that work internally with higher rank rotational tensors, e.g., vector displacements between atoms, and tensor products thereof. Here we present a framework for extending the Hierarchically Interacting Particle Neural Network (HIP-NN) with Tensor Sensitivity information (HIP-NN-TS) from each local atomic environment. Crucially, the method employs a weight tying strategy that allows direct incorporation of many-body information while adding very few model parameters. We show that HIP-NN-TS is more accurate than HIP-NN, with negligible increase in parameter count, for several datasets and network sizes. As the dataset becomes more complex, tensor sensitivities provide greater improvements to model accuracy. In particular, HIP-NN-TS achieves a record mean absolute error of 0.927 kcal/mol for conformational energy variation on the challenging COMP6 benchmark, which includes a broad set of organic molecules. We also compare the computational performance of HIP-NN-TS to HIP-NN and other models in the literature.

References in corpus (8)

• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• A Model of Inductive Bias Learning
• A Fourth-Generation High-Dimensional Neural Network Potential with Accurate Electrostatics Including Non-local Charge Transfer
• SpookyNet: Learning Force Fields with Electronic Degrees of Freedom and Nonlocal Effects
• ANI-1: A data set of 20M off-equilibrium DFT calculations for organic molecules
• Gaussian Moments as Physically Inspired Molecular Descriptors for Accurate and Scalable Machine Learning Potentials
• A Novel Approach to Describe Chemical Environments in High Dimensional Neural Network Potentials

Cited by in corpus (3)

• Machine Learning Framework for Modeling Exciton-Polaritons in Molecular Materials
• Machine-Learned Potentials for Solvation Modeling
• Knowledge Distillation of Noisy Force Labels for Improved Coarse-Grained Force Fields