BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260421T090514Z
LOCATION:Plenary Room (Bldg. 6 - 001)
DTSTART;TZID=Europe/Stockholm:20260629T194900
DTEND;TZID=Europe/Stockholm:20260629T195000
UID:submissions.pasc-conference.org_PASC26_sess124_pos103@linklings.com
SUMMARY:Parallel Tempering on Boundary Conditions with Normalizing Flows t
 o Solve Topological Freezing
DESCRIPTION:Victor Granados (University of Bern)\n\nIn particle physics, L
 attice Quantum Chromodynamics (LQCD) studies the strong interaction, respo
 nsible, for example, for the binding of atomic nuclei, through computation
 al methods.\nAn essential part of LQCD consists on being able to sample hi
 gh-dimensional multi-modal distributions, for which direct sampling method
 s are not available.\nStandard methods based on Markov Chain Monte Carlo a
 lgorithms have proven useful, but face short-comings such as long autocorr
 elations, often due to being unable to sample distributions where regions 
 of high probability are separated by long regions of low probability, a pr
 oblem known as Topological Freezing in LQCD.\nIn this work, we explore a s
 olution to Topological Freezing with Parallel Tempering on Boundary Condit
 ions (PTBC) and normalizing flows.\nThe former is an algorithm that allows
  traveling through low-probability regions by evolving several Markov chai
 ns in parallel with slightly different conditions and proposing exchanges 
 between the different chains.\nNormalizing flows are a machine learning me
 thod that can learn how to generate samples for a complex distribution sta
 rting with samples from an easier distribution.\nBy accelerating the PTBC 
 algorithm with normalizing flows, we aim to obtain samples with lower auto
 correlations.\n\n
END:VEVENT
END:VCALENDAR