PoS - Proceedings of Science
Volume 449 - The European Physical Society Conference on High Energy Physics (EPS-HEP2023) - T10 Searches for New Physics
Unsupervised tagging of semivisible jets with normalized autoencoders in CMS
F. Eble*  on behalf of the CMS Collaboration
Full text: pdf
Pre-published on: January 24, 2024
Published on: March 21, 2024
A particularly interesting application of autoencoders (AE) for High Energy Physics is their use as anomaly detection (AD) algorithms to perform a signal-agnostic search for new physics. This is achieved by training the AE on standard model physics and tagging potential signal events as anomalies. The use of an AE as an AD algorithm relies on the assumption that the network better reconstructs examples it was trained on than ones drawn from a different probability distribution, i.e. anomalies. Using the search for non resonant production of semivisible jets as a benchmark, we demonstrate the tendency of AEs to generalize beyond the dataset they are trained on, hindering their performance. We show how normalized AEs, specifically designed to suppress this effect, give a sizable boost in performance. We further propose a different loss function and signal-agnostic training stopping condition to reach the optimal performance.
DOI: https://doi.org/10.22323/1.449.0491
How to cite

Metadata are provided both in "article" format (very similar to INSPIRE) as this helps creating very compact bibliographies which can be beneficial to authors and readers, and in "proceeding" format which is more detailed and complete.

Open Access
Creative Commons LicenseCopyright owned by the author(s) under the term of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.