Speaker
David Albandea
(University of Valencia - IFIC)
Description
The recent introduction of machine learning tecniques, especially normalizing flows, for the sampling of lattice gauge theories has shed some hope on improving the sampling efficiency of the traditional HMC algorithm. However, naive usage of normalizing flows has been shown to lead to bad scaling with the volume. In this talk we propose using local normalizing flows at a scale given by the correlation length. Even if naively these transformations have a very small acceptance, when combined with the HMC lead to algorithms with high acceptance and reduced autocorrelation times compared with HMC. Several scaling tests are performed in the $\phi^{4}$ theory in 2D.
Primary authors
Dr
Alberto Ramos
(University of Valencia - IFIC)
David Albandea
(University of Valencia - IFIC)
Joe Marsh Rossney
(University of Edinburgh)
Luigi Del Debbio
(University of Edinburgh)
Prof.
Pilar Hernandez
(University of Valencia - IFIC)
Prof.
Richard Kenway
(University of Edinburgh)
