Inference of cosmic-ray source properties by conditional invertible neural networks

• Verfasst von: Bister, Teresa [VerfasserIn]
• Verfasst von: Erdmann, Martin [VerfasserIn]
• Verfasst von: Köthe, Ullrich [VerfasserIn]
• Verfasst von: Schulte, Josina [VerfasserIn]
• Titel: Inference of cosmic-ray source properties by conditional invertible neural networks
• Verf.angabe: Teresa Bister, Martin Erdmann, Ullrich Köthe, Josina Schulte
• E-Jahr: 2022
• Jahr: 24 February 2022
• Umfang: 10 S.
• Fussnoten: Gesehen am 03.05.2022
• Titel Quelle: Enthalten in: The European physical journal. C, Particles and fields
• Ort Quelle: Berlin : Springer, 1998
• Jahr Quelle: 2022
• Band/Heft Quelle: 82(2022), Artikel-ID 171, Seite 1-10
• ISSN Quelle: 1434-6052
• DOI: doi:10.1140/epjc/s10052-022-10138-x
• Datenträger: Online-Ressource
• Sprache: eng
• K10plus-PPN: 180076751X

Abstract

The inference of physical parameters from measured distributions constitutes a core task in physics data analyses. Among recent deep learning methods, so-called conditional invertible neural networks provide an elegant approach owing to their probability-preserving bijective mapping properties. They enable training the parameter-observation correspondence in one mapping direction and evaluating the parameter posterior distributions in the reverse direction. Here, we study the inference of cosmic-ray source properties from cosmic-ray observations on Earth using extensive astrophysical simulations. We compare the performance of conditional invertible neural networks (cINNs) with the frequently used Markov Chain Monte Carlo (MCMC) method. While cINNs are trained to directly predict the parameters’ posterior distributions, the MCMC method extracts the posterior distributions through a likelihood function that matches simulations with observations. Overall, we find good agreement between the physics parameters derived by the two different methods. As a result of its computational efficiency, the cINN method allows for a swift assessment of inference quality.