BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Interpretable Model-Independent Detection of New Physics Signals  
 - Purvasha Chakravarti (Imperial College London)
DTSTART:20220225T140000Z
DTEND:20220225T150000Z
UID:TALK168116@talks.cam.ac.uk
CONTACT:Qingyuan Zhao
DESCRIPTION:A central goal in experimental high energy physics is to detec
 t new physics signals that are not explained by known physics. In this tal
 k I will present our algorithm that we use to search for new signals that 
 appear as deviations from known Standard Model physics in high-dimensional
  particle physics data. To do this\, we determine whether there is any sta
 tistically significant difference between the distribution of Standard Mod
 el background samples and the distribution of the experimental observation
 s\, which are a mixture of the background and a potential new signal. We d
 o this without making any model assumptions on the signal that we are sear
 ching for.  We use a classifier and construct three test statistics using 
 the classifier: an estimated likelihood ratio test (LRT) statistic\, a tes
 t based on the area under the ROC curve (AUC)\, and a test based on the mi
 sclassification error (MCE)\, to detect the presence of the signal in the 
 experimental data. Additionally\, I will present our methods for estimatin
 g the signal strength parameter and interpreting the high-dimensional clas
 sifier in order to understand the properties of the detected signal. I wil
 l also present the results on a data set related to the search for the Hig
 gs boson at the Large Hadron Collider at CERN. 
LOCATION:MR12\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR