Classification of Fermi-LAT unassociated sources with machine learning in the presence of dataset shifts

Malyshev D (2025)


Publication Type: Conference contribution

Publication year: 2025

Journal

Publisher: EDP Sciences

Book Volume: 319

Conference Proceedings Title: EPJ Web of Conferences

Event location: Roma, ITA

DOI: 10.1051/epjconf/202531906009

Abstract

About one third of Fermi-LAT sources are unassociated. We perform multi-class classification of Fermi-LAT sources using machine learning with the goal of probabilistic classification of the unassociated sources. A particular attention is paid to the fact that the distributions of associated and unassociated sources are different as functions of source parameters. In this work, we address this problem in the framework of dataset shifts in machine learning.

Authors with CRIS profile

How to cite

APA:

Malyshev, D. (2025). Classification of Fermi-LAT unassociated sources with machine learning in the presence of dataset shifts. In Antonio Capone, Silvia Celli, Claudio Gasbarra, Aldo Morselli (Eds.), EPJ Web of Conferences. Roma, ITA: EDP Sciences.

MLA:

Malyshev, Dmitry. "Classification of Fermi-LAT unassociated sources with machine learning in the presence of dataset shifts." Proceedings of the 9th Roma International Conference on Astroparticle Physics, RICAP 2024, Roma, ITA Ed. Antonio Capone, Silvia Celli, Claudio Gasbarra, Aldo Morselli, EDP Sciences, 2025.

BibTeX: Download