Shared Data and Algorithms for Deep Learning in Fundamental Physics

Computing and Software for Big Science, vol. 6

Abstract

Abstract We introduce a Python package that provides simple and unified access to a collection of datasets from fundamental physics research—including particle physics, astroparticle physics, and hadron- and nuclear physics—for supervised machine learning studies. The datasets contain hadronic top quarks, cosmic-ray-induced air showers, phase transitions in hadronic matter, and generator-level histories. While public datasets from multiple fundamental physics disciplines already exist, the common interface and provided reference models simplify future work on cross-disciplinary machine learning and transfer learning in fundamental physics. We discuss the design and structure and line out how additional datasets can be submitted for inclusion. As showcase application, we present a simple yet flexible graph-based neural network architecture that can easily be applied to a wide range of supervised learning tasks. We show that our approach reaches performance close to dedicated methods on all datasets. To simplify adaptation for various problems, we provide easy-to-follow instructions on how graph-based representations of data structures, relevant for fundamental physics, can be constructed and provide code implementations for several of them. Implementations are also provided for our proposed method and all reference algorithms.

Authors 13

1. L. Benato

   Universität Hamburg

   Affiliation as printed

   Institut für Experimentalphysik, Universität Hamburg, Hamburg, Germany

2. Erik Buhmann

   Universität Hamburg

   Affiliation as printed

   Institut für Experimentalphysik, Universität Hamburg, Hamburg, Germany

3. Martin Erdmann Aachen III. Physikalisches Institut A

   RWTH Aachen University

   Affiliation as printed

   III. Physikalisches Institut A, RWTH Aachen University, Aachen, Germany

4. P. Fackeldey Aachen III. Physikalisches Institut A

   RWTH Aachen University

   Affiliation as printed

   III. Physikalisches Institut A, RWTH Aachen University, Aachen, Germany

5. Jonas Glombitza Aachen III. Physikalisches Institut A

   RWTH Aachen University

   Affiliation as printed

   III. Physikalisches Institut A, RWTH Aachen University, Aachen, Germany

6. N. M. Hartmann

   Ludwig-Maximilians-Universität München

   Affiliation as printed

   Fakultät für Physik, Ludwig Maximilians University Munich, Munich, Germany

7. Gregor Kasieczka corresponding

   Universität Hamburg

   Affiliation as printed

   Institut für Experimentalphysik, Universität Hamburg, Hamburg, Germany

8. W. Korcari

   Universität Hamburg

   Affiliation as printed

   Institut für Experimentalphysik, Universität Hamburg, Hamburg, Germany

9. T. Kuhr

   Ludwig-Maximilians-Universität München

   Affiliation as printed

   Fakultät für Physik, Ludwig Maximilians University Munich, Munich, Germany

10. Jan Steinheimer

    Frankfurt Institute for Advanced Studies

    Affiliation as printed

    Frankfurt Institute for Advanced Studies (FIAS), Frankfurt am Main, Germany

11. HORST STÖCKER

    Frankfurt Institute for Advanced Studies

    Affiliation as printed

    Frankfurt Institute for Advanced Studies (FIAS), Frankfurt am Main, Germany

12. Tilman Plehn

    Heidelberg University

    Affiliation as printed

    Institut für Theoretische Physik, Universität Heidelberg, Heidelberg, Germany

13. Kai Zhou

    Frankfurt Institute for Advanced Studies

    Affiliation as printed

    Frankfurt Institute for Advanced Studies (FIAS), Frankfurt am Main, Germany