SciPost logo

SciPost Submission Page

Beyond Cuts in Small Signal Scenarios -- Enhanced Sneutrino Detectability Using Machine Learning

by Daniel Alvestad, Nikolai Fomin, Jörn Kersten, Steffen Maeland, Inga Strümke

Submission summary

As Contributors: Daniel Alvestad · Nikolai Fomin · Jörn Kersten · Steffen Maeland · Inga Strümke
Arxiv Link: https://arxiv.org/abs/2108.03125v2 (pdf)
Date submitted: 2021-12-27 07:06
Submitted by: Kersten, Jörn
Submitted to: SciPost Physics Core
Academic field: Physics
Specialties:
  • High-Energy Physics - Phenomenology
Approaches: Theoretical, Computational, Phenomenological

Abstract

We investigate enhancing the sensitivity of new physics searches at the LHC by machine learning in the case of background dominance and a high degree of overlap between the observables for signal and background. We use two different models, XGBoost and a deep neural network, to exploit correlations between observables and compare this approach to the traditional cut-and-count method. We consider different methods to analyze the models' output, finding that a template fit generally performs better than a simple cut. By means of a Shapley decomposition, we gain additional insight into the relationship between event kinematics and the machine learning model output. We consider a supersymmetric scenario with a metastable sneutrino as a concrete example, but the methodology can be applied to a much wider class of supersymmetric models.

Current status:
Editor-in-charge assigned


Author comments upon resubmission

We would like to thank the referees for the detailed comments and suggestions. We have prepared a detailed PDF document with our reply to each point. It is available at https://filesender.uninett.no/?s=download&token=425affed-8f3d-4d2b-84c0-3c1dd839f172 (file RefereeReply_v1.pdf).

List of changes

We have listed major changes in the reply to the referees mentioned above. In addition, we have made minor changes for increased readability throughout the manuscript. A PDF document showing all changes is available at https://filesender.uninett.no/?s=download&token=425affed-8f3d-4d2b-84c0-3c1dd839f172 (file diff.pdf).

Submission & Refereeing History

You are currently on this page

Resubmission 2108.03125v2 on 27 December 2021

Login to report or comment