Generalised Dynamic Radius Jets for Robust Collider Analyses
En palabras de los autores
Jets and their reconstructions play a central role in precision measurements and searches for new physics at hadron colliders. Conventional jet clustering algorithms employ a fixed radius parameter, which may not optimally describe events containing jets of varying characteristic sizes. Building on the recently proposed dynamic radius jet clustering framework, we construct several substructure-inspired dynamic radius prescriptions based on jet angularities and energy correlation functions. A detailed study of these algorithms is performed, including detector effects within the Delphes framework and in the presence of high pileup corresponding to an average of 150 interactions per event, with pileup contamination mitigated using the PUPPI algorithm. The performance of the proposed algorithms is compared with that of the standard anti- algorithm in boosted () and events against dijet backgrounds. Using jet substructure observables and multivariate analysis based on boosted decision trees, we find that the dynamic radius algorithms lead to improved reconstruction of boosted heavy-particle jets and achieve better signal-background discrimination compared to the conventional fixed radius anti- clustering.
Apareció: viernes, 25 de septiembre. arXiv. Preprint, todavía sin revisión por pares.
Comentario de los autores: 33 pages, 10 figures, and 4 tables