Deep Learning of Mean First Passage Time Scape: Chemical Short-Range Order and Kinetics of Diffusive Relaxation

TL;DR

This paper introduces a deep learning framework that models atomic-scale diffusive relaxation over long timescales, enabling the study of slow processes like phase transitions with high accuracy and computational efficiency.

Contribution

It presents a novel method combining deep neural networks and reinforcement learning to predict mean first passage times in atomistic systems, extending simulation capabilities to longer timescales.

Findings

Accurately predicts disorder-to-order transition timescales in CrCoNi alloy

Bridges the gap between atomistic simulation and experimental measurements

Extends modeling of slow diffusive processes to previously inaccessible timescales

Abstract

Processes slow compared to atomic vibrations pose significant challenges in atomistic simulations, particularly for phenomena such as diffusive relaxations and phase transitions, where repeated crossings and the shear number of thermally activated transitions make direct numerical simulations impossible. We present a computational framework that captures atomic-scale diffusive relaxation over extended timescales by learning the mean first passage time (MFPT) with a deep neural network. The model is trained via a self-consistent recursive formulation based on the Markovian assumption, relying solely on local residence times and transition probabilities between neighboring states. Furthermore, we leverage deep reinforcement learning (DRL)-accelerated atomistic simulations to expedite the identification of thermodynamic equilibrium and the generation of accurate physical transition probabilities. Applied to vacancy-mediated chemical short-range order (SRO) evolution in equiatomic CrCoNi, our method uncovers disorder-to-order transition timescales in quantitative agreement with experimental measurements. By bridging the gap between simulation and experiment, our approach extends atomistic modeling to previously inaccessible timescales and offers a predictive tool for navigating process-structure-property relationships.