BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Extracting the most from collider data with deep learning - Benjam
 in Nachman (LBNL)
DTSTART:20200519T150000Z
DTEND:20200519T160000Z
UID:TALK142396@talks.cam.ac.uk
CONTACT:William Fawcett
DESCRIPTION:Precise scientific analysis in collider-based particle physics
  is possible because of complex simulations that connect fundamental theor
 ies to observable quantities. These simulations have been paired with mult
 ivariate methods for many years in search of the smallest distance scales 
 in nature. Deep learning tools hold great promise to qualitatively change 
 this paradigm by allowing for holistic analysis of data in its natural hyp
 erdimensionality with thousands or millions of features instead of up to t
 ens of features. These tools are not yet broadly used for all areas of dat
 a analysis because of the traditional dependence on simulations. In this t
 alk\, I will discuss how we can change this paradigm in order to exploit t
 he new features of deep learning to explore nature at sub-nuclear distance
  scales. In particular\, I will show how neural networks can be used to (1
 ) overcome the challenge of intractable hypervariate probability density m
 odelling and (2) learn directly from (unlabelled) data to perform\nhypothe
 sis tests that go beyond any existing analysis methods. The example for (1
 ) will be full phase space unfolding and the example for (2) will be anoma
 ly detection. The talk will include a discussion of uncertainties associat
 ed with deep learning-based analyses.
LOCATION:Ryle Seminar Room (Rutherford 930) 
END:VEVENT
END:VCALENDAR