mlsanityData to 7 Oct 2026

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

arXiv 2406.00249 ↗First posted 1 Jun 2024cs.LG cs.CRcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 30 of its 53 references that are in the map. First posted on arXiv 1 Jun 2024.

Builds on

30 of 53 references.

References · 30

  1. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  2. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
  3. 2016Deep Learning with Differential Privacycited by 2,682
  4. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
  5. 2016Federated Learning: Strategies for Improving Communication Efficiencycited by 1,461
  6. 2019Deep Leakage from Gradientscited by 1,004
  7. 2018On First-Order Meta-Learning Algorithmscited by 1,003
  8. 2020Meta-Learning in Neural Networks: A Surveycited by 944
  9. 2020Inverting Gradients -- How easy is it to break privacy in federated learning?cited by 707
  10. 2016Stealing Machine Learning Models via Prediction APIscited by 602
  11. 2018Exploiting Unintended Feature Leakage in Collaborative Learningcited by 509
  12. 2018Knockoff Nets: Stealing Functionality of Black-Box Modelscited by 285
18 more
  1. 2020Systematic Evaluation of Privacy Risks of Machine Learning Modelscited by 273
  2. 2019Stolen Memories: Leveraging Model Memorization for Calibrated White-Box Membership Inferencecited by 168
  3. 2020A Survey of Privacy Attacks in Machine Learningcited by 143
  4. 2022Reconstructing Training Data with Informed Adversariescited by 143
  5. 2019Privacy Risks of Securing Machine Learning Models against Adversarial Examplescited by 136
  6. 2013Hacking Smart Machines with Smarter Ones: How to Extract Meaningful Data from Machine Learning Classifierscited by 128
  7. 2021When the Curious Abandon Honesty: Federated Learning Is Not Privatecited by 126
  8. 2019MeLU: Meta-Learned User Preference Estimator for Cold-Start Recommendationcited by 119
  9. 2021Robbing the Fed: Directly Obtaining Private Data in Federated Learning with Modified Modelscited by 106
  10. 2018Model Reconstruction from Model Explanationscited by 84
  11. 2018Understanding Membership Inferences on Well-Generalized Learning Modelscited by 83
  12. 2020Privacy in Deep Learning: A Surveycited by 70
  13. 2018Copycat CNN: Stealing Knowledge by Persuading Confession with Random Non-Labeled Datacited by 69
  14. 2019Differentially Private Meta-Learningcited by 40
  15. 2016You are Who You Know and How You Behave: Attribute Inference Attacks via Users' Social Friends and Behaviorscited by 34
  16. 2021Task-adaptive Neural Process for User Cold-Start Recommendationcited by 32
  17. 2020MetaSelector: Meta-Learning for Recommendation with User-Level Adaptive Model Selectioncited by 13
  18. 2022How to Combine Membership-Inference Attacks on Multiple Updated Modelscited by 8

Built on by

Not cited yet.