Enhancing Learning with Label Differential Privacy by Vector Approximation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhao, Puning, Fan, Rongfei, Wu, Huiwen, Li, Qingming, Wu, Jiafei, Liu, Zhe
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929357356793856
author Zhao, Puning
Fan, Rongfei
Wu, Huiwen
Li, Qingming
Wu, Jiafei
Liu, Zhe
author_facet Zhao, Puning
Fan, Rongfei
Wu, Huiwen
Li, Qingming
Wu, Jiafei
Liu, Zhe
contents Label differential privacy (DP) is a framework that protects the privacy of labels in training datasets, while the feature vectors are public. Existing approaches protect the privacy of labels by flipping them randomly, and then train a model to make the output approximate the privatized label. However, as the number of classes $K$ increases, stronger randomization is needed, thus the performances of these methods become significantly worse. In this paper, we propose a vector approximation approach, which is easy to implement and introduces little additional computational overhead. Instead of flipping each label into a single scalar, our method converts each label into a random vector with $K$ components, whose expectations reflect class conditional probabilities. Intuitively, vector approximation retains more information than scalar labels. A brief theoretical analysis shows that the performance of our method only decays slightly with $K$. Finally, we conduct experiments on both synthesized and real datasets, which validate our theoretical analysis as well as the practical performance of our method.
format Preprint
id arxiv_https___arxiv_org_abs_2405_15150
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Enhancing Learning with Label Differential Privacy by Vector Approximation
Zhao, Puning
Fan, Rongfei
Wu, Huiwen
Li, Qingming
Wu, Jiafei
Liu, Zhe
Machine Learning
Label differential privacy (DP) is a framework that protects the privacy of labels in training datasets, while the feature vectors are public. Existing approaches protect the privacy of labels by flipping them randomly, and then train a model to make the output approximate the privatized label. However, as the number of classes $K$ increases, stronger randomization is needed, thus the performances of these methods become significantly worse. In this paper, we propose a vector approximation approach, which is easy to implement and introduces little additional computational overhead. Instead of flipping each label into a single scalar, our method converts each label into a random vector with $K$ components, whose expectations reflect class conditional probabilities. Intuitively, vector approximation retains more information than scalar labels. A brief theoretical analysis shows that the performance of our method only decays slightly with $K$. Finally, we conduct experiments on both synthesized and real datasets, which validate our theoretical analysis as well as the practical performance of our method.
title Enhancing Learning with Label Differential Privacy by Vector Approximation
topic Machine Learning
url https://arxiv.org/abs/2405.15150