EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

How to understand limitations of generative networks

Ranit Das, Luigi Favaro, Theo Heimel, Claudius Krause, Tilman Plehn, David Shih

SciPost Physics · 25.01.2024 · 16 (1)

DOI: 10.21468/scipostphys.16.1.031 ↗

Kurz gefasst

Fachartikel in der Zeitschrift SciPost Physics, online erschienen am 25.01.2024. Verfasst von Ranit Das und 5 weiteren Autor:innen. Die Arbeit wurde bisher 42-mal zitiert (Crossref) und nennt 83 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Baden-Württemberg Stiftung, Carl-Zeiss-Stiftung, Deutsche Forschungsgemeinschaft.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Well-trained classifiers and their complete weight distributions provide us with a well-motivated and practicable method to test generative networks in particle physics. We illustrate their benefits for distribution-shifted jets, calorimeter showers, and reconstruction-level events. In all cases, the classifier weights make for a powerful test of the generative network, identify potential problems in the density estimation, relate them to the underlying physics, and tie in with a comprehensive precision and uncertainty treatment for generative networks.

Förderung

  • Baden-Württemberg Stiftung
  • Carl-Zeiss-Stiftung
  • Deutsche Forschungsgemeinschaft
  • United States Department of Energy

Zitieren

APA

Das, R., Favaro, L., Heimel, T., Krause, C., Plehn, T., & Shih, D. (2024). How to understand limitations of generative networks. SciPost Physics, 16(1). https://doi.org/10.21468/scipostphys.16.1.031

Vancouver

Das R, Favaro L, Heimel T, Krause C, Plehn T, Shih D. How to understand limitations of generative networks. SciPost Physics. 2024;16(1). doi:10.21468/scipostphys.16.1.031

BibTeX
@article{Das2024,
  author = {Das, Ranit and Favaro, Luigi and Heimel, Theo and Krause, Claudius and Plehn, Tilman and Shih, David},
  title = {How to understand limitations of generative networks},
  journal = {SciPost Physics},
  year = {2024},
  volume = {16},
  number = {1},
  doi = {10.21468/scipostphys.16.1.031}
}
Kostenlos starten →