Equivariant Flow-Based Sampling for Lattice Gauge Theory

Overview

Journal Phys Rev Lett

Publisher American Physical Society

Specialty Biophysics

Date 2020 Oct 5

PMID 33016765

Citations 6

Authors

Gurtej Kanwar

Michael S Albergo

Denis Boyda

Kyle Cranmer

Daniel C Hackett

Sebastien Racaniere

Danilo Jimenez Rezende

Phiala E Shanahan

Affiliations

Soon will be listed here.

Abstract

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, at small bare coupling, 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.

Citing Articles

Foundations of automatic feature extraction at LHC-point clouds and graphs.

Bhardwaj A, Konar P, Ngairangbam V Eur Phys J Spec Top. 2024; 233(15-16):2619-2640.

PMID: 39605978 PMC: 11588817. DOI: 10.1140/epjs/s11734-024-01306-z.

Sampling with flows, diffusion, and autoregressive neural networks from a spin-glass perspective.

Ghio D, Dandi Y, Krzakala F, Zdeborova L Proc Natl Acad Sci U S A. 2024; 121(27):e2311810121.

PMID: 38913892 PMC: 11228464. DOI: 10.1073/pnas.2311810121.

Scientific discovery in the age of artificial intelligence.

Wang H, Fu T, Du Y, Gao W, Huang K, Liu Z Nature. 2023; 620(7972):47-60.

PMID: 37532811 DOI: 10.1038/s41586-023-06221-2.

Non-Hertz-Millis scaling of the antiferromagnetic quantum critical metal via scalable Hybrid Monte Carlo.

Lunts P, Albergo M, Lindsey M Nat Commun. 2023; 14(1):2547.

PMID: 37137882 PMC: 10156689. DOI: 10.1038/s41467-023-37686-4.

Algorithmic Differentiation for Automated Modeling of Machine Learned Force Fields.

Schmitz N, Muller K, Chmiela S J Phys Chem Lett. 2022; 13(43):10183-10189.

PMID: 36279418 PMC: 9639201. DOI: 10.1021/acs.jpclett.2c02632.