2021/07/31 by Jiawei Yan, Hugo Touchette, Grant M. Rotskoff · 37 citations
Neuroscience · Physics and Astronomy · #Advanced Thermodynamics and Statistical Mechanics #Classical mechanics #Computer science #Dynamical system (definition) #Dynamical systems theory #Function (biology) #Neural dynamics and brain function #Non-equilibrium thermodynamics #Phase (matter) #Phase space #Phase transition #Physics #Quantum mechanics #Representation (politics) #Sampling (signal processing) #Statistical Mechanics and Entropy #Statistical physics #Trajectory #cond-mat.stat-mech
paper · pdf · doi:10.1103/physreve.105.024115
published in Physical review. E 105(2), 024115 (American Physical Society) · 11 pages, 5 figures. v2: corrected version, close to published version
openalex publication_date 2022/02/09 · arxiv created 2022/02/11 · arxiv updated 2022/02/14 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
Sampling the collective, dynamical fluctuations that lead to nonequilibrium pattern formation requires probing rare regions of trajectory space. Recent approaches to this problem, based on importance sampling, cloning, and spectral approximations, have yielded significant insight into nonequilibrium systems but tend to scale poorly with the size of the system, especially near dynamical phase transitions. Here we propose a machine learning algorithm that samples rare trajectories and estimates the associated large deviation functions using a many-body control force by leveraging the flexible function representation provided by deep neural networks, importance sampling in trajectory space, and stochastic optimal control theory. We show that this approach scales to hundreds of interacting particles and remains robust at dynamical phase transitions.