mlsanityData to 7 Oct 2026

Understanding Data Reconstruction Leakage in Federated Learning from a Theoretical Perspective

arXiv 2408.12119 ↗First posted 22 Aug 2024cs.CR cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Federated learning (FL) is an emerging collaborative learning paradigm that aims to protect data privacy. Unfortunately, recent works show FL algorithms are vulnerable to the serious data reconstruction attacks. However, existing works lack a theoretical foundation on to what extent the devices' data can be reconstructed and the effectiveness of these attacks cannot be compared fairly due to their unstable performance. To address this deficiency, we propose a theoretical framework to understand data reconstruction attacks to FL. Our framework involves bounding the data reconstruction error and an attack's error bound reflects its inherent attack effectiveness. Under the framework, we can theoretically compare the effectiveness of existing attacks. For instance, our results on multiple datasets validate that the iDLG attack inherently outperforms the DLG attack.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 39 of its 50 references that are in the map. First posted on arXiv 22 Aug 2024.

Builds on

39 of 50 references.

References · 39

  1. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  2. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
  3. 2018Federated Optimization in Heterogeneous Networkscited by 2,971
  4. 2016Deep Learning with Differential Privacycited by 2,682
  5. 2019Federated Learning: Challenges, Methods, and Future Directionscited by 1,531
  6. 2019On the Convergence of FedAvg on Non-IID Datacited by 1,220
  7. 2019Deep Leakage from Gradientscited by 1,004
  8. 2019Towards Federated Learning at Scale: System Designcited by 826
  9. 2020Inverting Gradients -- How easy is it to break privacy in federated learning?cited by 707
  10. 2020The Future of Digital Health with Federated Learningcited by 625
  11. 2018Local SGD Converges Fast and Communicates Littlecited by 595
  12. 2019Federated Learning with Differential Privacy: Algorithms and Performance Analysiscited by 528
27 more
  1. 2018Lipschitz regularity of deep neural networks: analysis and efficient estimationcited by 413
  2. 2019Algorithm Unrolling: Interpretable, Efficient Deep Learning for Signal and Image Processingcited by 411
  3. 2018Sparsified SGD with Memorycited by 373
  4. 2020Label-Only Membership Inference Attackscited by 371
  5. 2020iDLG: Improved Deep Leakage from Gradientscited by 367
  6. 2017Deep Models Under the GAN: Information Leakage from Collaborative Deep Learningcited by 326
  7. 2021See through Gradients: Image Batch Recovery via GradInversioncited by 320
  8. 2018Parallel Restarted SGD with Faster Convergence and Less Communication: Demystifying Why Model Averaging Works for Deep Learningcited by 277
  9. 2020Systematic Evaluation of Privacy Risks of Machine Learning Modelscited by 273
  10. 2018Beyond Inferring Class Representatives: User-Level Privacy Leakage From Federated Learningcited by 242
  11. 2022Reconstructing Training Data from Trained Neural Networkscited by 129
  12. 2021Robbing the Fed: Directly Obtaining Private Data in Federated Learning with Modified Modelscited by 106
  13. 2021Gradient Inversion with Generative Image Priorcited by 101
  14. 2020Neural Networks are Convex Regularizers: Exact Polynomial-time Convex Optimization Formulations for Two-layer Networkscited by 96
  15. 2020R-GAP: Recursive Gradient Attack on Privacycited by 95
  16. 2020Provable Defense against Privacy Leakage in Federated Learning from Representation Perspectivecited by 94
  17. 2018Theoretical Linear Convergence of Unfolded ISTA and its Practical Weights and Thresholdscited by 83
  18. 2022Fishing for User Data in Large-Batch Federated Learning via Gradient Magnificationcited by 72
  19. 2022Auditing Privacy Defenses in Federated Learning via Generative Gradient Leakagecited by 72
  20. 2022Bounding Training Data Reconstruction in Private (Deep) Learningcited by 43
  21. 2023Bounding Training Data Reconstruction in DP-SGDcited by 42
  22. 2021Bayesian Framework for Gradient Leakagecited by 34
  23. 2020Privacy-preserving Collaborative Learning with Automatic Transformation Searchcited by 31
  24. 2022Learning to Invert: Simple Adaptive Attacks for Gradient Inversion in Federated Learningcited by 30
  25. 2021Revealing and Protecting Labels in Distributed Trainingcited by 17
  26. 2019An Algorithm Unrolling Approach to Deep Image Deblurringcited by 14
  27. 2024Inf2Guard: An Information-Theoretic Framework for Learning Privacy-Preserving Representations against Inference Attackscited by 13

Built on by

Not cited yet.