Machine Learning the Entropy to Estimate Free Energy Differences without Sampling Transitions

TL;DR

This paper introduces a deep learning-based method to estimate free energy differences between metastable states using only short, separate simulations, avoiding the need for transition sampling or prior knowledge of slow modes.

Contribution

The authors develop a novel deep learning approach to estimate entropy and free energy from separate phase simulations, enabling accurate free energy calculations without transition sampling.

Findings

Achieves state-of-the-art accuracy in melting temperature estimation for Na and Al.

Does not require prior knowledge of transition pathways or slow modes.

Successfully benchmarks on crystalline and liquid metals.

Abstract

Thermodynamic phase transitions, a central concept in physics and chemistry, are typically controlled by an interplay of enthalpic and entropic contributions. In most cases, the estimation of the enthalpy in simulations is straightforward but evaluating the entropy is notoriously hard. As a result, it is common to induce transitions between the metastable states and estimate their relative occupancies, from which the free energy difference can be inferred. However, for systems with large free energy barriers, sampling these transitions is a significant computational challenge. Dedicated enhanced sampling algorithms require significant prior knowledge of the slow modes governing the transition, which is typically unavailable. We present an alternative approach, which only uses short simulations of each phase separately. We achieve this by employing a recently developed deep learning model for estimating the entropy and hence the free energy of each metastable state. We benchmark our approach calculating the free energies of crystalline and liquid metals. Our method features state-of-the-art precision in estimating the melting transition temperature in Na and Al without requiring any prior information or simulation of the transition pathway itself.