统计物理学
模拟退火
应用数学
朗之万动力
玻尔兹曼常数
数学优化
马尔可夫过程
梯度下降
数学
计算机科学
物理
人工神经网络
统计
热力学
机器学习
出处
期刊:Cornell University - arXiv
日期:2024-01-01
标识
DOI:10.48550/arxiv.2401.05553
摘要
Probably one of the most striking examples of the close connections between global optimization processes and statistical physics is the simulated annealing method, inspired by the famous Monte Carlo algorithm devised by Metropolis et al. in the middle of the last century. In this paper we show how the tools of linear kinetic theory allow to describe this gradient-free algorithm from the perspective of statistical physics and how convergence to the global minimum can be related to classical entropy inequalities. This analysis highlight the strong link between linear Boltzmann equations and stochastic optimization methods governed by Markov processes. Thanks to this formalism we can establish the connections between the simulated annealing process and the corresponding mean-field Langevin dynamics characterized by a stochastic gradient descent approach. Generalizations to other selection strategies in simulated annealing that avoid the acceptance-rejection dynamic are also provided.
科研通智能强力驱动
Strongly Powered by AbleSci AI