Learning New Physics from an Imperfect Machine
Overview
Affiliations
We show how to deal with uncertainties on the Standard Model predictions in an agnostic new physics search strategy that exploits artificial neural networks. Our approach builds directly on the specific Maximum Likelihood ratio treatment of uncertainties as nuisance parameters for hypothesis testing that is routinely employed in high-energy physics. After presenting the conceptual foundations of our method, we first illustrate all aspects of its implementation and extensively study its performances on a toy one-dimensional problem. We then show how to implement it in a multivariate setup by studying the impact of two typical sources of experimental uncertainties in two-body final states at the LHC.
Nanosecond anomaly detection with decision trees and real-time application to exotic Higgs decays.
Roche S, Bayer Q, Carlson B, Ouligian W, Serhiayenka P, Stelzer J Nat Commun. 2024; 15(1):3527.
PMID: 38664390 PMC: 11045859. DOI: 10.1038/s41467-024-47704-8.
Quantum anomaly detection for collider physics.
Alvi S, Bauer C, Nachman B J High Energy Phys. 2023; 2023(2):220.
PMID: 36852337 PMC: 9946862. DOI: 10.1007/JHEP02(2023)220.
Learning new physics efficiently with nonparametric methods.
Letizia M, Losapio G, Rando M, Grosso G, Wulzer A, Pierini M Eur Phys J C Part Fields. 2022; 82(10):879.
PMID: 36212113 PMC: 9534824. DOI: 10.1140/epjc/s10052-022-10830-y.
Unsupervised Quark/Gluon Jet Tagging With Poissonian Mixture Models.
Alvarez E, Spannowsky M, Szewc M Front Artif Intell. 2022; 5:852970.
PMID: 35372834 PMC: 8969742. DOI: 10.3389/frai.2022.852970.