Improving the performance of weak supervision searches using data augmentation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Zong-En, Chiang, Cheng-Wei, Hsieh, Feng-Yang
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917875299647488
author Chen, Zong-En
Chiang, Cheng-Wei
Hsieh, Feng-Yang
author_facet Chen, Zong-En
Chiang, Cheng-Wei
Hsieh, Feng-Yang
contents Weak supervision combines the advantages of training on real data with the ability to exploit signal properties. However, training a neural network using weak supervision often requires an excessive amount of signal data, which severely limits its practical applicability. In this study, we propose addressing this limitation through data augmentation, increasing the training data's size and diversity. Specifically, we focus on physics-inspired data augmentation methods, such as $p_{\text{T}}$ smearing and jet rotation. Our results demonstrate that data augmentation can significantly enhance the performance of weak supervision, enabling neural networks to learn efficiently from substantially less data.
format Preprint
id arxiv_https___arxiv_org_abs_2412_00198
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Improving the performance of weak supervision searches using data augmentation
Chen, Zong-En
Chiang, Cheng-Wei
Hsieh, Feng-Yang
High Energy Physics - Phenomenology
Machine Learning
High Energy Physics - Experiment
Weak supervision combines the advantages of training on real data with the ability to exploit signal properties. However, training a neural network using weak supervision often requires an excessive amount of signal data, which severely limits its practical applicability. In this study, we propose addressing this limitation through data augmentation, increasing the training data's size and diversity. Specifically, we focus on physics-inspired data augmentation methods, such as $p_{\text{T}}$ smearing and jet rotation. Our results demonstrate that data augmentation can significantly enhance the performance of weak supervision, enabling neural networks to learn efficiently from substantially less data.
title Improving the performance of weak supervision searches using data augmentation
topic High Energy Physics - Phenomenology
Machine Learning
High Energy Physics - Experiment
url https://arxiv.org/abs/2412.00198