EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Anomalies, representations, and self-supervision

Barry M. Dillon, Luigi Favaro, Friedrich Feiden, Tanmoy Modak, Tilman Plehn

SciPost Physics Core · 16.08.2024 · 7 (3)

DOI: 10.21468/scipostphyscore.7.3.056 ↗

Kurz gefasst

Fachartikel in der Zeitschrift SciPost Physics Core, online erschienen am 16.08.2024. Verfasst von Barry M. Dillon und 4 weiteren Autor:innen. Die Arbeit wurde bisher 17-mal zitiert (Crossref) und nennt 69 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Alexander von Humboldt-Stiftung, Deutsche Forschungsgemeinschaft.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We develop a self-supervised method for density-based anomaly detection using contrastive learning, and test it using event-level anomaly data from CMS ADC2021. The AnomalyCLR technique is data-driven and uses augmentations of the background data to mimic non-Standard-Model events in a model-agnostic way. It uses a permutation-invariant Transformer Encoder architecture to map the objects measured in a collider event to the representation space, where the data augmentations define a representation space which is sensitive to potential anomalous features. An AutoEncoder trained on background representations then computes anomaly scores for a variety of signals in the representation space. With AnomalyCLR we find significant improvements on performance metrics for all signals when compared to the raw data baseline.

Förderung

  • Alexander von Humboldt-Stiftung
  • Deutsche Forschungsgemeinschaft

Zitieren

APA

Dillon, B. M., Favaro, L., Feiden, F., Modak, T., & Plehn, T. (2024). Anomalies, representations, and self-supervision. SciPost Physics Core, 7(3). https://doi.org/10.21468/scipostphyscore.7.3.056

Vancouver

Dillon BM, Favaro L, Feiden F, Modak T, Plehn T. Anomalies, representations, and self-supervision. SciPost Physics Core. 2024;7(3). doi:10.21468/scipostphyscore.7.3.056

BibTeX
@article{Dillon2024,
  author = {Dillon, Barry M. and Favaro, Luigi and Feiden, Friedrich and Modak, Tanmoy and Plehn, Tilman},
  title = {Anomalies, representations, and self-supervision},
  journal = {SciPost Physics Core},
  year = {2024},
  volume = {7},
  number = {3},
  doi = {10.21468/scipostphyscore.7.3.056}
}
Kostenlos starten →