Abstract:The accuracy of atomistic simulations depends on the precision of force fields. Traditional numerical methods often struggle to optimize the empirical force field parameters for reproducing target properties. Recent approaches rely on training these force fields based on forces and energies from first-principle simulations. However, it is unclear whether these approaches will enable capturing complex material responses such as vibrational, or elastic properties. To this extent, we introduce a framework, employing inner loop simulations and outer loop optimization, that exploits automatic differentiation for both property prediction and force-field optimization by computing gradients of the simulation analytically. We demonstrate the approach by optimizing classical Stillinger-Weber and EDIP potentials for silicon systems to reproduce the elastic constants, vibrational density of states, and phonon dispersion. We also demonstrate how a machine-learned potential can be fine-tuned using automatic differentiation to reproduce any target property such as radial distribution functions. Interestingly, the resulting force field exhibits improved accuracy and generalizability to unseen temperatures than those fine-tuned on energies and forces. Finally, we demonstrate the extension of the approach to optimize the force fields towards multiple target properties. Altogether, differentiable simulations, through the analytical computation of their gradients, offer a powerful tool for both theoretical exploration and practical applications toward understanding physical systems and materials.

Reversible molecular simulation for training classical and machine learning force fields

Differentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins

Towards Exact Molecular Dynamics Simulations with Machine-Learned Force Fields

Differentiable simulation to develop molecular dynamics force fields for disordered proteins

Molecular Dynamics with Neural-Network Potentials

A transferable active-learning strategy for reactive molecular force fields

Molecular Force Fields with Gradient-Domain Machine Learning: Construction and Application to Dynamics of Small Molecules with Coupled Cluster Forces

Molecular Dynamics with On-the-Fly Machine Learning of Quantum-Mechanical Forces

Differentiable Molecular Simulations for Control and Learning

Machine Learning of coarse-grained Molecular Dynamics Force Fields

Molecular Force Fields with Gradient-Domain Machine Learning (GDML): Comparison and Synergies with Classical Force Fields

On-the-fly active learning of interpretable Bayesian force fields for atomistic rare events

Force field optimization by end-to-end differentiable atomistic simulation

Learning from the Density to Correct Total Energy and Forces in First Principle Simulations

Efficient Machine Learning Force Field for Large-Scale Molecular Simulations of Organic Systems

Machine Learning Directed Optimization of Classical Molecular Modeling Force Fields

chemtrain: Learning Deep Potential Models via Automatic Differentiation and Statistical Physics

Synthetic Force-Field Database for Training Machine Learning Models to Predict Mobility-Preserving Coarse-Grained Molecular-Simulation Potentials

Reinforcement Learning for Multi-Scale Molecular Modeling

Top-down machine learning of coarse-grained protein force-fields