Atomistic learning in the electronically grand-canonical ensemble

A strategy is presented for the machine-learning emulation of electronic structure calculations carried out in the electronically grand-canonical ensemble. The approach relies upon a dual-learning scheme, where both the system charge and the system energy are predicted for each image. The scheme is shown to be capable of emulating basic electrochemical reactions at a range of potentials, and coupling it with a bootstrap-ensemble approach gives reasonable estimates of the prediction uncertainty. The method is also demonstrated to accelerate saddle-point searches, and to extrapolate to systems with one to five water layers. We anticipate that this method will allow for larger length- and time-scale simulations necessary for electrochemical simulations.

Automated summary

This article presents a strategy for machine-learning emulation of electronic structure calculations in the electronically grand-canonical ensemble. The approach uses a dual-learning scheme to predict both system charge and system energy for each image. The scheme has been shown to successfully emulate basic electrochemical reactions at various potentials and combining it with a bootstrap-ensemble approach gives reasonable estimates of prediction uncertainty. The method also accelerates saddle-point searches and extrapolates to systems with different numbers of water layers. This method is expected to enable larger length- and time-scale simulations necessary for electrochemical simulations.