arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2003.06413hep-latcond-mat.stat-mechcs.LG
收藏

格点规范理论的等变流采样

Equivariant flow-based sampling for lattice gauge theory

Massachusetts Institute of Technology(麻省理工学院);New York University(纽约大学);DeepMind(深度思维)

Gurtej Kanwar, Michael S. Albergo, Denis Boyda, Kyle Cranmer, Daniel C. Hackett, Sébastien Racanière, Danilo Jimenez Rezende, Phiala E. Shanahan

摘要:

我们定义了一类机器学习的基于流的采样算法的格点规范理论,规范不变的建设。我们证明了这个框架的应用U(1)规范理论在两个时空维度,并发现在参数空间的临界点附近的方法是数量级更有效的采样拓扑量比更传统的采样程序,如混合蒙特卡罗和热浴。

英文摘要:

We define a class of machine-learned flow-based sampling algorithms for lattice gauge theories that are gauge-invariant by construction. We demonstrate the application of this framework to U(1) gauge theory in two spacetime dimensions, and find that near critical points in parameter space the approach is orders of magnitude more efficient at sampling topological quantities than more traditional sampling procedures such as Hybrid Monte Carlo and Heat Bath.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑