Testing trivializing maps in the Hybrid Monte Carlo algorithm

2011/02/09 by Georg P. Engel, Stefan Schaefer · 1 citation
Physics and Astronomy · #hep-lat

paper · pdf · doi:10.1016/j.cpc.2011.05.004

published as Comput.Phys.Commun.182:2107-2114,2011 · 18 pages, 7 figures

arxiv created 2011/02/09 · arxiv updated 2015/03/18

Abstract

We test a recent proposal to use approximate trivializing maps in a field theory to speed up Hybrid Monte Carlo simulations. Simulating the CP^N-1 model, we find a small improvement with the leading order transformation, which is however compensated by the additional computational overhead. The scaling of the algorithm towards the continuum is not changed. In particular, the effect of the topological modes on the autocorrelation times is studied.

Cited by

• Lecture Notes on Normalizing Flows for Lattice Quantum Field Theories

Related

• Learning Trivializing Flows
• Towards reduction of autocorrelation in HMC by machine learning
• Machine Learning Trivializing Maps: A First Step Towards Understanding How Flow-Based Samplers Scale Up
• Learning Trivializing Flows in a ϕ⁴ theory from coarser lattices
• Improvement in Autocorrelation Times Measured by the Master-Field Technique using Field Transformation HMC in 2+1 Domain Wall Fermion Simulations