mlsanityData to 14 Oct 2026

Provable Defense against Privacy Leakage in Federated Learning from Representation Perspective

arXiv 2012.06043 ↗First posted 8 Dec 2020cs.LG cs.AI cs.CVcited by 94

Not on the map. Why?

Abstract (arXiv)

Federated learning (FL) is a popular distributed learning framework that can reduce privacy risks by not explicitly sharing private data. However, recent works demonstrated that sharing model updates makes FL vulnerable to inference attacks. In this work, we show our key observation that the data representation leakage from gradients is the essential cause of privacy leakage in FL. We also provide an analysis of this observation to explain how the data presentation is leaked. Based on this observation, we propose a defense against model inversion attack in FL. The key idea of our defense is learning to perturb data representation such that the quality of the reconstructed data is severely degraded, while FL performance is maintained. In addition, we derive certified robustness guarantee to FL and convergence guarantee to FedAvg, after applying our defense. To evaluate our defense, we conduct experiments on MNIST and CIFAR10 for defending against the DLG attack and GS attack. Without sacrificing accuracy, the results demonstrate that our proposed defense can increase the mean squared error between the reconstructed data and the raw data by as much as more than 160X for both DLG attack and GS attack, compared with baseline defense methods. The privacy of the FL system is significantly improved.

Key facts. Cited by 94 papers in the map (data through 14 Oct 2026). First posted on arXiv 8 Dec 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 6; 2022: 18; 2023: 24; 2024: 19; 2025: 17; 2026: 10.

94 papers.

Sort:

By area:

  1. 2023Towards Building the Federated GPT: Federated Instruction Tuningcited by 185
  2. 2023Heterogeneous Federated Learning: State-of-the-art and Research Challengescited by 161
  3. 2021When the Curious Abandon Honesty: Federated Learning Is Not Privatecited by 127
  4. 2021Robbing the Fed: Directly Obtaining Private Data in Federated Learning with Modified Modelscited by 108
  5. 2022Survey on Federated Learning Threats: concepts, taxonomy on attacks and defences, experimental study and challengescited by 84
  6. 2022Auditing Privacy Defenses in Federated Learning via Generative Gradient Leakagecited by 72
  7. 2022LAMP: Extracting Text from Gradients with Language Model Priorscited by 54
  8. 2023GIFD: A Generative Gradient Inversion Method with Feature Domain Optimizationcited by 45
  9. 2022Do Gradient Inversion Attacks Make Federated Learning Unsafe?cited by 44
  10. 2022A Survey on Gradient Inversion: Attacks, Defenses and Future Directionscited by 40
  11. 2022Gradient Obfuscation Gives a False Sense of Security in Federated Learningcited by 39
  12. 2021Bayesian Framework for Gradient Leakagecited by 34
38 more
  1. 2022CD2-pFed: Cyclic Distillation-guided Channel Decoupling for Model Personalization in Federated Learningcited by 34
  2. 2022Reconstructing Training Data from Model Gradient, Provablycited by 27
  3. 2022Data Leakage in Federated Averagingcited by 26
  4. 2023FedBPT: Efficient Federated Black-box Prompt Tuning for Large Language Modelscited by 26
  5. 2023The Best of Both Worlds: Accurate Global and Personalized Models through Federated Learning with Data-Free Hyper-Knowledge Distillationcited by 25
  6. 2022FedMix: Mixed Supervised Federated Learning for Medical Image Segmentationcited by 23
  7. 2023A Survey of What to Share in Federated Learning: Perspectives on Model Utility, Privacy Leakage, and Communication Efficiencycited by 21
  8. 2021Federated Split Vision Transformer for COVID-19 CXR Diagnosis using Task-Agnostic Trainingcited by 17
  9. 2022Distributed Contrastive Learning for Medical Image Segmentationcited by 16
  10. 2023On the Trustworthiness Landscape of State-of-the-art Generative Models: A Survey and Outlookcited by 15
  11. 2024Trustworthy Distributed AI Systems: Robustness, Privacy, and Governancecited by 14
  12. 2024GI-NAS: Boosting Gradient Inversion Attacks Through Adaptive Neural Architecture Searchcited by 14
  13. 2021Federated Deep Learning with Bayesian Privacycited by 13
  14. 2023Privacy Assessment on Reconstructed Images: Are Existing Evaluation Metrics Faithful to Human Perception?cited by 13
  15. 2024Inf2Guard: An Information-Theoretic Framework for Learning Privacy-Preserving Representations against Inference Attackscited by 13
  16. 2022Federated Learning in Non-IID Settings Aided by Differentially Private Synthetic Datacited by 12
  17. 2021SPATL: Salient Parameter Aggregation and Transfer Learning for Heterogeneous Clients in Federated Learningcited by 11
  18. 2022Dropout is NOT All You Need to Prevent Gradient Leakagecited by 11
  19. 2022Concealing Sensitive Samples against Gradient Leakage in Federated Learningcited by 11
  20. 2023A Survey on Vulnerability of Federated Learning: A Learning Algorithm Perspectivecited by 9
  21. 2023Robust and IP-Protecting Vertical Federated Learning against Unexpected Quitting of Partiescited by 7
  22. 2024Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attackscited by 7
  23. 2022FedDef: Defense Against Gradient Leakage in Federated Learning-based Network Intrusion Detection Systemscited by 6
  24. 2023On the Robustness of Split Learning against Adversarial Attackscited by 6
  25. 2023Understanding Deep Gradient Leakage via Inversion Influence Functionscited by 6
  26. 2024A New Federated Learning Framework Against Gradient Inversion Attackscited by 6
  27. 2025CENSOR: Defense Against Gradient Inversion via Orthogonal Subspace Bayesian Samplingcited by 5
  28. 2023FCA: Taming Long-tailed Federated Medical Image Classification by Classifier Anchoringcited by 4
  29. 2023Gradient Leakage Defense with Key-Lock Module for Federated Learningcited by 4
  30. 2023FedPDD: A Privacy-preserving Double Distillation Framework for Cross-silo Federated Recommendationcited by 4
  31. 2023Temporal Gradient Inversion Attacks with Robust Optimizationcited by 4
  32. 2023PrivaScissors: Enhance the Privacy of Collaborative Inference through the Lens of Mutual Informationcited by 4
  33. 2023Artificial Intelligence for Web 3.0: A Comprehensive Surveycited by 4
  34. 2023Maximum Knowledge Orthogonality Reconstruction with Gradients in Federated Learningcited by 4
  35. 2023PPIDSG: A Privacy-Preserving Image Distribution Sharing Scheme with GAN in Federated Learningcited by 4
  36. 2024Analyzing Inference Privacy Risks Through Gradients in Machine Learningcited by 4
  37. 2024Seeing the Forest through the Trees: Data Leakage from Partial Transformer Gradientscited by 3
  38. 2022Defense Against Gradient Leakage Attacks via Learning to Obscure Datacited by 2

50 most cited of 94 All 94 (JSON)