Jiming Yang | Searcharxiv

arXiv2023

Detecting subtle macroscopic changes in a finite temperature classical scalar field with machine learning

The ability to detect macroscopic changes is important for probing the behaviors of experimental many-body systems from the classical to the quantum realm. Although abrupt changes near phase boundaries can easily be detected, subtle macroscopic changes are much more difficult to detect as the changes can be obscured by noise. In this study, as a toy model for detecting subtle macroscopic changes in many-body systems, we try to differentiate scalar field samples at varying temperatures. We compare different methods for making such differentiations, from physics method, statistics method, to AI method. Our finding suggests that the AI method outperforms both the statistical method and the physics method in its sensitivity. Our result provides a proof-of-concept that AI can potentially detect macroscopic changes in many-body systems that elude physical measures.

cond-mat.stat-mech↗

arXiv2022

Transition from localized to delocalized trajectories in random walk subject to random drives

Random walk subject to random drive has been extensively employed as a model for physical and biological processes. While equilibrium statistical physics has yielded significant insights into the distributions of dynamical fixed points of such a system, its non-equilibrium properties remain largely unexplored. In contrast, most real-world applications concern the dynamical aspects of this model. In particular, dynamical quantities like heat dissipation and work absorption play a central role in predicting and controlling non-equilibrium phases of matter. Recent advances in non-equilibrium statistical physics enable a more refined study of the dynamical aspects of random walk under random drives. We perform a numerical study on this model and demonstrate that it exhibits two distinct phases: a localized phase where typical random walk trajectories are non-extensive and confined to the neighborhood of fixed points, and a delocalized phase where typical random walk trajectories are extensive and can transition between fixed points. We propose different summary statistics for the heat dissipation and show that these two phases are distinctly different. Our characterization of these distinctive phases deepens the understanding of and provides novel strategies for the non-equilibrium phase of this model.

cond-mat.stat-mech↗