Accuracy Improvement in Differentially Private Logistic Regression: A Pre-training Approach

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Hoseinpour, Mohammad, Hoseinpour, Milad, Aghagolzadeh, Ali
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916120414388224
author Hoseinpour, Mohammad
Hoseinpour, Milad
Aghagolzadeh, Ali
author_facet Hoseinpour, Mohammad
Hoseinpour, Milad
Aghagolzadeh, Ali
contents Machine learning (ML) models can memorize training datasets. As a result, training ML models over private datasets can lead to the violation of individuals' privacy. Differential privacy (DP) is a rigorous privacy notion to preserve the privacy of underlying training datasets. Yet, training ML models in a DP framework usually degrades the accuracy of ML models. This paper aims to boost the accuracy of a DP logistic regression (LR) via a pre-training module. In more detail, we initially pre-train our LR model on a public training dataset that there is no privacy concern about it. Then, we fine-tune our DP-LR model with the private dataset. In the numerical results, we show that adding a pre-training module significantly improves the accuracy of the DP-LR model.
format Preprint
id arxiv_https___arxiv_org_abs_2307_13771
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Accuracy Improvement in Differentially Private Logistic Regression: A Pre-training Approach
Hoseinpour, Mohammad
Hoseinpour, Milad
Aghagolzadeh, Ali
Machine Learning
Cryptography and Security
Machine learning (ML) models can memorize training datasets. As a result, training ML models over private datasets can lead to the violation of individuals' privacy. Differential privacy (DP) is a rigorous privacy notion to preserve the privacy of underlying training datasets. Yet, training ML models in a DP framework usually degrades the accuracy of ML models. This paper aims to boost the accuracy of a DP logistic regression (LR) via a pre-training module. In more detail, we initially pre-train our LR model on a public training dataset that there is no privacy concern about it. Then, we fine-tune our DP-LR model with the private dataset. In the numerical results, we show that adding a pre-training module significantly improves the accuracy of the DP-LR model.
title Accuracy Improvement in Differentially Private Logistic Regression: A Pre-training Approach
topic Machine Learning
Cryptography and Security
url https://arxiv.org/abs/2307.13771