BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Event Generation and Statistical Sampling with Deep Generative Mod
els
DTSTART;VALUE=DATE-TIME:20190417T094500Z
DTEND;VALUE=DATE-TIME:20190417T100500Z
DTSTAMP;VALUE=DATE-TIME:20210117T131939Z
UID:indico-contribution-3357995@indico.cern.ch
DESCRIPTION:Speakers: Sydney Otten (Radboud Universiteit Nijmegen)\nWe pre
sent a study for the generation of events from a physical process with gen
erative deep learning. To simulate physical processes it is not only impor
tant to produce physical events\, but also to produce the events with the
right frequency of occurrence (density). We investigate the feasibility to
learn the event generation and the frequency of occurrence with Generativ
e Adversarial Networks (GANs) and Variational Autoencoders (VAEs) to produ
ce events like Monte Carlo generators. We study three toy models from high
energy physics\, i.e. a simple two-body decay\, the processes $e^+e^-\\to
Z \\to l^+l^-$ and $p p \\to t\\bar{t} $ including the decay of the top q
uarks and a simulation of the detector response. We show that GANs and the
standard VAE do not produce the right distributions. By buffering density
information of Monte Carlo events in latent space given the encoder of a
VAE we are able to construct a prior for the sampling of new events from t
he decoder that yields distributions that are in very good agreement with
real Monte Carlo events and are generated $\\mathcal{O}(10^8)$ times faste
r. Applications of this work include generic density estimation and sampli
ng\, targeted event generation via a principal component analysis of encod
ed events in the latent space and the possibility to generate better rando
m numbers for importance sampling\, e.g. for the phase space integration o
f matrix elements in quantum perturbation theories. The method also allows
to build event generators directly from real data events.\n\nhttps://indi
co.cern.ch/event/766872/contributions/3357995/
LOCATION:CERN 500/1-001 - Main Auditorium
URL:https://indico.cern.ch/event/766872/contributions/3357995/
END:VEVENT
END:VCALENDAR