Multiple testing for signal-agnostic searches of new physics with machine learning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909293524025344 |
|---|---|
| author | Grosso, Gaia Letizia, Marco |
| author_facet | Grosso, Gaia Letizia, Marco |
| contents | In this work, we address the question of how to enhance signal-agnostic searches by leveraging multiple testing strategies. Specifically, we consider hypothesis tests relying on machine learning, where model selection can introduce a bias towards specific families of new physics signals. We show that it is beneficial to combine different tests, characterised by distinct choices of hyperparameters, and that performances comparable to the best available test are generally achieved while providing a more uniform response to various types of anomalies. Focusing on the New Physics Learning Machine, a methodology to perform a signal-agnostic likelihood-ratio test, we explore a number of approaches to multiple testing, such as combining p-values and aggregating test statistics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_12296 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Multiple testing for signal-agnostic searches of new physics with machine learning Grosso, Gaia Letizia, Marco High Energy Physics - Phenomenology Machine Learning High Energy Physics - Experiment Data Analysis, Statistics and Probability Methodology In this work, we address the question of how to enhance signal-agnostic searches by leveraging multiple testing strategies. Specifically, we consider hypothesis tests relying on machine learning, where model selection can introduce a bias towards specific families of new physics signals. We show that it is beneficial to combine different tests, characterised by distinct choices of hyperparameters, and that performances comparable to the best available test are generally achieved while providing a more uniform response to various types of anomalies. Focusing on the New Physics Learning Machine, a methodology to perform a signal-agnostic likelihood-ratio test, we explore a number of approaches to multiple testing, such as combining p-values and aggregating test statistics. |
| title | Multiple testing for signal-agnostic searches of new physics with machine learning |
| topic | High Energy Physics - Phenomenology Machine Learning High Energy Physics - Experiment Data Analysis, Statistics and Probability Methodology |
| url | https://arxiv.org/abs/2408.12296 |