Jan 15 – 17, 2024
Institute of Nuclear Physics PAN, Cracow Poland
Europe/Warsaw timezone

Machine Learning @CREDO

Jan 17, 2024, 9:50 AM
30m
Institute of Nuclear Physics PAN, Cracow Poland

Institute of Nuclear Physics PAN, Cracow Poland

Speaker

Łukasz Bibrzycki (AGH University of Krakow)

Description

The recent, ongoing as well as planned CREDO ML research activities will be discussed. So far CREDO ML focused on the analysis of individual cosmic ray events including the filtering of artifacts and hardware noise and attepmts to associate event shapes with particular particle types. In terms of ML methodologies these were supervised binary and multi-class problems, respectively. We also applied non-supervised schemes to extract the most representative features of events or isolate outliers and novelties. The projects recently undertaken focus on generative models which enable the modelling of collective cosmic ray phenomena like Extensive Air Showers, in particular studying particle distributions or unfolding detector effects.

Primary author

Łukasz Bibrzycki (AGH University of Krakow)

Co-authors

Marcin Piekarczyk (AGH University of Krakow) Michał Niedźwiecki (Cracow University of Technology) Olaf Bar (Cracow University of Technology) Sławomir Stuglik (Institute of Nuclear Physics PAS) Tomasz Hachaj (AGH University of Krakow)

Presentation materials