IML Machine Learning Working Group: unsupervised searches and unfolding with ML

Europe/Zurich
40-S2-A01 - Salle Anderson (CERN)

40-S2-A01 - Salle Anderson

CERN

100
Show room on map
Videoconference Rooms
IML-MachineLearning-WG
Name
IML-MachineLearning-WG
Description
Vidyo room for the IML machine learning working group
Extension
10453344
Owner
Steven Schramm
Auto-join URL
Useful links
Phone numbers
    • 15:00 15:10
      Introduction and news 10m
      Speakers: Lorenzo Moneta (CERN), Markus Stoye (Imperial College (GB)), Paul Seyfert (CERN), Rudiger Haake (Yale University (US)), Steven Schramm (Universite de Geneve (CH))
    • 15:10 15:20
      ML community white paper path forward 10m
      Speaker: Dr Sergei Gleyzer (University of Florida (US))
    • 15:25 15:55
      Guiding New Physics Searches with Unsupervised Learning 30m

      I will describe an approach to search for new phenomena in data, by detecting discrepancies between two datasets. These could be, for example, a simulated standard-model background, and an observed dataset containing a potential hidden signal of New Physics.
      I will propose a new statistical test, built upon a test statistic which measures deviations between two samples, using a Nearest Neighbors approach to estimate the local ratio of the density of points.
      The test is model-independent and non-parametric, requiring no knowledge of the shape of the underlying distributions, and it does not bin the data, thus retaining full information from the multidimensional feature space.
      As a by-product, the technique is also a useful tool to identify regions of interest for further study.
      As a proof-of-concept, I will show the power of the method when applied to synthetic Gaussian data, and to a simulated dark matter signal at the LHC.

      Speaker: Andrea De Simone (SISSA)
    • 16:00 16:30
      Learning New Physics from a machine 30m

      We propose using neural networks to detect data departures from a given reference model, with no prior bias on the nature of the new physics responsible for the discrepancy. The model-independent nature of our approach, and its ability to deal with rare signals such as those expected at the LHC, is quantitatively assessed in toy examples.

      Speaker: Andrea Wulzer (CERN)
    • 16:35 16:55
      Machine learning as an instrument for data unfolding 20m
      Speaker: Alexander Glazov (Deutsches Elektronen-Synchrotron (DE))
Your browser is out of date!

Update your browser to view this website correctly. Update my browser now

×