Speaker
Ivan Kharuk
(INR)
Description
The mass composition of ultra-high-energy cosmic rays can be analyzed by employing deep neural networks. We present an improved version of such analysis for Telescope Array's surface detectors data. Our neural network was trained on a large Monte-Carlo dataset simulating the expected experimental data distribution, and then was applied to the actual experimental data. Systematic and model errors are discussed.
Keywords
machine learning; neural networks; mass composition; ultra-high-energy cosmic rays
| Subcategory | Experimental Results |
|---|---|
| Collaboration | Telescope Array |
Primary authors
Ivan Kharuk
(INR)
Mikhail Kuznetsov
(INR RAS, Moscow)
Yana Zhezher
(ICRR, University of Tokyo & INR RAS, Moscow)
Grigory Rubtsov
(Institute for Nuclear Research of the Russian Academy of Sciences)
Oleg Kalashev
(INR RAS)
For the Telescope Array Collaboration
