BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:GPU Implementation of Bayesian Neural Networks in SUSY Studies
DTSTART;VALUE=DATE-TIME:20131014T130000Z
DTEND;VALUE=DATE-TIME:20131014T134500Z
DTSTAMP;VALUE=DATE-TIME:20200229T045610Z
UID:indico-contribution-1512667@indico.cern.ch
DESCRIPTION:Speakers: Michelle Perry (Florida State University)\nThe searc
h for new physics has typically been guided by theoretical models with rel
atively few parameters. However\, recently\, more general models\, such as
the 19-parameter phenomenological minimal supersymmetric standard model (
pMSSM)\, have been used to interpret data at the Large Hadron Collider. Un
fortunately\, due to the complexity of the calculations\, the predictions
of these models are available at a discrete set of parameter points\, whic
h makes the use of analysis techniques that require smooth maps between th
e parameters and a given prediction problematic. It would be useful\, ther
efore\, to have a computationally routine way to construct such mappings.
We propose to construct the mappings using Bayesian neural networks (BNN).
Bayesian neural networks have been used in a few high-profile analyses i
n high energy physics for both classification and functional approximation
. The main limitation to their widespread use is the time required to con
struct these functions. In this talk\, we describe an efficient Graphical
Processing Unit (GPU) implementation of the construction of BNNs using the
Hybrid Markov-Chain Monte Carlo (MCMC) method. We describe our implementa
tion of the MCMC algorithm on the GPU\, including the speedups we have ach
ieved so far and illustrate the effectiveness of our implementation by map
ping the pMSSM parameter space to some of its key predictions.\n\nhttps://
indico.cern.ch/event/214784/contributions/1512667/
LOCATION:Amsterdam\, Beurs van Berlage Grote zaal
URL:https://indico.cern.ch/event/214784/contributions/1512667/
END:VEVENT
END:VCALENDAR