Jun 5 – 9, 2023
Genova, Italy
Europe/Rome timezone

Machine Learning exotic hadrons

Jun 8, 2023, 5:25 PM
DAD - Room Benvenuto (Genova, Italy)

DAD - Room Benvenuto

Genova, Italy

Invited Analysis tools Analysis tools


Łukasz Bibrzycki (Pedagogical University of Krakow)


A densely connected feed-forward neural network is capable to classify poles of scattering matrix if fed with experimentally measured values of energy-dependent production intensity. As shown in [1], such a neural network trained with synthetic differential intensities calculated with scattering length approximated amplitudes classifies the $P_c$(4312) signal as a virtual state located at the 4th Riemann sheet with very high certainty. This is in line with the results of other analyses but surpasses them by providing the simultaneous evaluation of probabilities of competing scenarios, like eg. the interpretation in terms of the bound state. Studying the dimensionally reduced training and inference data obtained with the Principal Component Analysis gives us a certainty that our physical interpretation is robust. Moreover, using the Shapley Additive Explanations we can identify the energy bins which are key for the physical interpretation.

1. Deep Learning Exotic Hadrons, JPAC Collaboration • L. Ng, Ł. Bibrzycki, J. Nys, C. Fernandez-Ramirez, A. Pilloni, V. Mathieu, A.J. Rasmusson, A.P. Szczepaniak, Phys.Rev.D 105 (2022) 9, L091501

Primary authors

A J Rasmusson Adam Szczepaniak (Indiana University) Dr Alessandro Pilloni (UniME & CT) Cesar Fernandez-Ramirez (UNED/ICN-UNAM) Jannes Nys (Ghent University) Lawrence Ng (Florida State University) Vincent Mathieu Łukasz Bibrzycki (Pedagogical University of Krakow)

Presentation materials

Peer reviewing