Stress and heat flux via automatic differentiation
This provides a practical solution for materials scientists using advanced potentials, though it is incremental as it extends existing automatic differentiation methods to new outputs.
The authors tackled the problem of efficiently computing gradients like forces, stress, and heat flux for complex machine-learning potentials, demonstrating a unified automatic differentiation approach that was tested on Lennard-Jones and applied to predict cohesive properties and thermal conductivity of tin selenide.
Machine-learning potentials provide computationally efficient and accurate approximations of the Born-Oppenheimer potential energy surface. This potential determines many materials properties and simulation techniques usually require its gradients, in particular forces and stress for molecular dynamics, and heat flux for thermal transport properties. Recently developed potentials feature high body order and can include equivariant semi-local interactions through message-passing mechanisms. Due to their complex functional forms, they rely on automatic differentiation (AD), overcoming the need for manual implementations or finite-difference schemes to evaluate gradients. This study demonstrates a unified AD approach to obtain forces, stress, and heat flux for such potentials, and provides a model-independent implementation. The method is tested on the Lennard-Jones potential, and then applied to predict cohesive properties and thermal conductivity of tin selenide using an equivariant message-passing neural network potential.