Privacy Challenges in Meta-Learning: An Investigation on Model-Agnostic Meta-Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Rafiei, Mina, Maheri, Mohammadmahdi, Rabiee, Hamid R.
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910471726039040
author Rafiei, Mina
Maheri, Mohammadmahdi
Rabiee, Hamid R.
author_facet Rafiei, Mina
Maheri, Mohammadmahdi
Rabiee, Hamid R.
contents Meta-learning involves multiple learners, each dedicated to specific tasks, collaborating in a data-constrained setting. In current meta-learning methods, task learners locally learn models from sensitive data, termed support sets. These task learners subsequently share model-related information, such as gradients or loss values, which is computed using another part of the data termed query set, with a meta-learner. The meta-learner employs this information to update its meta-knowledge. Despite the absence of explicit data sharing, privacy concerns persist. This paper examines potential data leakage in a prominent metalearning algorithm, specifically Model-Agnostic Meta-Learning (MAML). In MAML, gradients are shared between the metalearner and task-learners. The primary objective is to scrutinize the gradient and the information it encompasses about the task dataset. Subsequently, we endeavor to propose membership inference attacks targeting the task dataset containing support and query sets. Finally, we explore various noise injection methods designed to safeguard the privacy of task data and thwart potential attacks. Experimental results demonstrate the effectiveness of these attacks on MAML and the efficacy of proper noise injection methods in countering them.
format Preprint
id arxiv_https___arxiv_org_abs_2406_00249
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Privacy Challenges in Meta-Learning: An Investigation on Model-Agnostic Meta-Learning
Rafiei, Mina
Maheri, Mohammadmahdi
Rabiee, Hamid R.
Machine Learning
Cryptography and Security
Meta-learning involves multiple learners, each dedicated to specific tasks, collaborating in a data-constrained setting. In current meta-learning methods, task learners locally learn models from sensitive data, termed support sets. These task learners subsequently share model-related information, such as gradients or loss values, which is computed using another part of the data termed query set, with a meta-learner. The meta-learner employs this information to update its meta-knowledge. Despite the absence of explicit data sharing, privacy concerns persist. This paper examines potential data leakage in a prominent metalearning algorithm, specifically Model-Agnostic Meta-Learning (MAML). In MAML, gradients are shared between the metalearner and task-learners. The primary objective is to scrutinize the gradient and the information it encompasses about the task dataset. Subsequently, we endeavor to propose membership inference attacks targeting the task dataset containing support and query sets. Finally, we explore various noise injection methods designed to safeguard the privacy of task data and thwart potential attacks. Experimental results demonstrate the effectiveness of these attacks on MAML and the efficacy of proper noise injection methods in countering them.
title Privacy Challenges in Meta-Learning: An Investigation on Model-Agnostic Meta-Learning
topic Machine Learning
Cryptography and Security
url https://arxiv.org/abs/2406.00249