8–13 Aug 2022
Hörsaalzentrum Poppelsdorf
Europe/Berlin timezone

Gauge-equivariant flow models for sampling in lattice field theories with pseudofermions

8 Aug 2022, 15:00
20m
CP1-HSZ/1.004 (CP1-HSZ) - HS7 (CP1-HSZ)

CP1-HSZ/1.004 (CP1-HSZ) - HS7

CP1-HSZ

70
Show room on map
Oral Presentation Algorithms (including Machine Learning, Quantum Computing, Tensor Networks) Algorithms

Speaker

Fernando Romero-Lopez (MIT)

Description

In this talk, we discuss gauge-equivariant architectures for flow-based sampling in fermionic lattice field theories with pseudofermions. We also discuss how flow-based sampling approaches can be improved by combination with standard techniques such as even/odd preconditioning and the Hasenbusch factorization. Numerical demonstrations in two-dimensional U(1) and SU(3) theories with $N_f=2$ flavors are provided.

Primary authors

Ryan Abbott (MIT) Michael Albergo (New York University) Denis Boyda (Argonne) Kyle Cranmer Dan Hackett Gurtej Kanwar (University of Bern) Sebastian Racaniere (Deep Mind) Danilo Rezende (Deep Mind) Fernando Romero-Lopez (MIT) Phiala Shanahan (MIT) Betsy Tian (MIT) Julian Urban (ITP Heidelberg)

Presentation materials