BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Evaluating ML-Accelerated Simulations of the Time Projection Chamb
 er for the MPD Experiment.
DTSTART;VALUE=DATE-TIME:20241025T144000Z
DTEND;VALUE=DATE-TIME:20241025T145500Z
DTSTAMP;VALUE=DATE-TIME:20260517T030521Z
UID:indico-contribution-4218@cern.ch
DESCRIPTION:Speakers: Fares Ghazzawi (HSE University\, Moscow\, Russia)\nC
 omputer-based simulations of high-energy physics experiments are\ncritical
  for obtaining more accurate physics results\, yet these simulations\ntend
  to be computationally expensive. Generative Machine Learning (ML) based \
 napproaches offer potential for accelerating the simulation for such exper
 iments. \nHowever\, a reduction in quality is often anticipated when compa
 ring these fast ML-based\nsimulations with detailed full simulations. In t
 his contribution\, we compare a \nML-based simulation to a detailed simula
 tion of the Time Projection Chamber\n(TPC) for the MPD experiment at the N
 ICA accelerator complex. We evaluate the \nextent to which high-level char
 acteristics\, such as the quality of reconstructed tracks\,\ncan and shoul
 d be reproduced by the ML-based fast simulation.\n\nhttps://indico.particl
 e.mephi.ru/event/436/contributions/4218/
LOCATION: Petrovskiy 2
URL:https://indico.particle.mephi.ru/event/436/contributions/4218/
END:VEVENT
END:VCALENDAR