KIT | KIT-Bibliothek | Impressum | Datenschutz

Punzi-loss and Punzi-net, outperforming standard MVA techniques in the search for new particles of unknown masses

Haigh, Huw; Feichtinger, Paul; Inguglia, Gianluca; Kahn, James ORCID iD icon


In this talk, we present the novel implementation of a non-differentiable metric approximation with a corresponding loss-scheduling based on the minimization of a figure-of-merit related function typical of particle physics (the so-called Punzi figure of merit). We call this new loss-scheduling a "Punzi-loss function" and the neural network that minimizes it a "Punzi-net". We tested the Punzi-net on simulated samples of signal and background at the Belle II experiment. We show that in the search for new particles of unknown mass, for example, a new Z’ boson, the Punzi-net outperforms standard multivariate analysis techniques and generalizes well to mass hypotheses for which it was not trained. This work constitutes a further step towards fully differentiable analyses in particle physics.

Zugehörige Institution(en) am KIT Steinbuch Centre for Computing (SCC)
Universität Karlsruhe (TH) – Zentrale Einrichtungen (Zentrale Einrichtungen)
Publikationstyp Poster
Publikationsjahr 2021
Sprache Englisch
Identifikator KITopen-ID: 1000140627
HGF-Programm 46.21.04 (POF IV, LK 01) HAICU
Veranstaltung 20th International Workshop on Advanced Computing and Analysis Techniques in Physics Research (ACAT 2021), Daejeon, Südkorea, 29.11.2021 – 03.12.2021
Relationen in KITopen
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page