Gamma/Hadron Separation in Imaging Air Cherenkov Telescopes Using Deep Learning Libraries TensorFlow and PyTorch. (arXiv:1811.11822v1 [astro-ph.IM])

Gamma/Hadron Separation in Imaging Air Cherenkov Telescopes Using Deep Learning Libraries TensorFlow and PyTorch. (arXiv:1811.11822v1 [astro-ph.IM])
<a href="http://arxiv.org/find/astro-ph/1/au:+Postnikov_E/0/1/0/all/0/1">E. B. Postnikov</a> (1), <a href="http://arxiv.org/find/astro-ph/1/au:+Kryukov_A/0/1/0/all/0/1">A. P. Kryukov</a> (1), <a href="http://arxiv.org/find/astro-ph/1/au:+Polyakov_S/0/1/0/all/0/1">S. P. Polyakov</a> (1), <a href="http://arxiv.org/find/astro-ph/1/au:+Shipilov_D/0/1/0/all/0/1">D. A. Shipilov</a> (2), <a href="http://arxiv.org/find/astro-ph/1/au:+Zhurov_D/0/1/0/all/0/1">D. P. Zhurov</a> (2) ((1) Lomonosov Moscow State University Skobeltsyn Institute of Nuclear Physics (MSU SINP), Moscow, Russia, (2) Irkutsk State University (ISU), Irkutsk, Russia)

In this work we compare two open source machine learning libraries, PyTorch
and TensorFlow, as software platforms for rejecting hadron background events
detected by imaging air Cherenkov telescopes (IACTs). Monte Carlo simulation
for the TAIGA-IACT telescope is used to estimate background rejection quality.
A wide variety of neural network algorithms provided by both libraries can
easily be tested on various types of data, which is useful for various imaging
air Cherenkov experiments. The work is a component of the Astroparticle.online
project, which collaborates with the TAIGA and KASCADE experiments and welcomes
any astroparticle experiment to join.

http://arxiv.org/icons/sfx.gif