mlsanityData to 7 Oct 2026

Fishing for User Data in Large-Batch Federated Learning via Gradient Magnification

arXiv 2202.00580 ↗First posted 1 Feb 2022cs.LG cs.CR cs.CVcited by 72

Not on the map. Why?

Abstract (arXiv)

Federated learning (FL) has rapidly risen in popularity due to its promise of privacy and efficiency. Previous works have exposed privacy vulnerabilities in the FL pipeline by recovering user data from gradient updates. However, existing attacks fail to address realistic settings because they either 1) require toy settings with very small batch sizes, or 2) require unrealistic and conspicuous architecture modifications. We introduce a new strategy that dramatically elevates existing attacks to operate on batches of arbitrarily large size, and without architectural modifications. Our model-agnostic strategy only requires modifications to the model parameters sent to the user, which is a realistic threat model in many scenarios. We demonstrate the strategy in challenging large-scale settings, obtaining high-fidelity data extraction in both cross-device and cross-silo federated learning.

Key facts. Cited by 72 papers in the map (data through 7 Oct 2026). Builds on 18 of its 37 references that are in the map. First posted on arXiv 1 Feb 2022.

Builds on

18 of 37 references.

References · 18

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  4. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  5. 2019Advances and Open Problems in Federated Learningcited by 3,021
  6. 2019Deep Leakage from Gradientscited by 1,004
  7. 2019Towards Federated Learning at Scale: System Designcited by 826
  8. 2020Inverting Gradients -- How easy is it to break privacy in federated learning?cited by 707
  9. 2021See through Gradients: Image Batch Recovery via GradInversioncited by 320
  10. 2019Differential Privacy Has Disparate Impact on Model Accuracycited by 316
  11. 2018Beyond Inferring Class Representatives: User-Level Privacy Leakage From Federated Learningcited by 242
  12. 2019Kornia: an Open Source Differentiable Computer Vision Library for PyTorchcited by 231
6 more
  1. 2021Evaluating Gradient Inversion Attacks and Defenses in Federated Learningcited by 196
  2. 2018Protection Against Reconstruction and Its Applications in Private Federated Learningcited by 127
  3. 2020R-GAP: Recursive Gradient Attack on Privacycited by 95
  4. 2021Eluding Secure Aggregation in Federated Learning via Model Inconsistencycited by 66
  5. 2020A Framework for Evaluating Gradient Leakage Attacks in Federated Learningcited by 63
  6. 2021Gradient Disaggregation: Breaking Privacy in Federated Learning by Reconstructing the User Participant Matrixcited by 43

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 13; 2023: 17; 2024: 21; 2025: 14; 2026: 7.

72 papers.

Sort:

By area:

  1. 2022Truth Serum: Poisoning Machine Learning Models to Reveal Their Secretscited by 94
  2. 2022Gradient Obfuscation Gives a False Sense of Security in Federated Learningcited by 37
  3. 2022On Privacy and Personalization in Cross-Silo Federated Learningcited by 36
  4. 2023LOKI: Large-scale Data Reconstruction Attack against Federated Learning through Model Manipulationcited by 35
  5. 2022Cocktail Party Attack: Breaking Aggregation-Based Privacy in Federated Learning using Independent Component Analysiscited by 29
  6. 2024Privacy Backdoors: Enhancing Membership Inference through Poisoning Pre-trained Modelscited by 28
  7. 2022Data Leakage in Federated Averagingcited by 26
  8. 2022CANIFE: Crafting Canaries for Empirical Privacy Measurement in Federated Learningcited by 24
  9. 2023Deconstructing Data Reconstruction: Multiclass, Weight Decay and General Lossescited by 24
  10. 2022On the (In)security of Peer-to-Peer Decentralized Machine Learningcited by 23
  11. 2024Privacy Backdoors: Stealing Data with Corrupted Pretrained Modelscited by 20
  12. 2023Scale-MIA: A Scalable Model Inversion Attack against Secure Federated Learning via Latent Space Reconstructioncited by 18
38 more
  1. 2023Reconstructing Individual Data Points in Federated Learning Hardened with Differential Privacy and Secure Aggregationcited by 16
  2. 2024SPEAR:Exact Gradient Inversion of Batches in Federated Learningcited by 14
  3. 2022TabLeak: Tabular Data Leakage in Federated Learningcited by 13
  4. 2024The Federation Strikes Back: A Survey of Federated Learning Privacy Attacks, Defenses, Applications, and Policy Landscapecited by 13
  5. 2024DAGER: Exact Gradient Inversion for Large Language Modelscited by 12
  6. 2022Privacy-preserving Decentralized Federated Learning over Time-varying Communication Graphcited by 11
  7. 2024Reconstructing Training Data From Real World Models Trained with Transfer Learningcited by 10
  8. 2023Backdoor Attacks in Peer-to-Peer Federated Learningcited by 9
  9. 2023Client-specific Property Inference against Secure Aggregation in Federated Learningcited by 9
  10. 2023The Resource Problem of Using Linear Layer Leakage Attack in Federated Learningcited by 9
  11. 2023Hiding in Plain Sight: Disguising Data Stealing Attacks in Federated Learningcited by 9
  12. 2025Exploring the Vulnerabilities of Federated Learning: A Deep Dive into Gradient Inversion Attackscited by 9
  13. 2022ScionFL: Efficient and Robust Secure Quantized Aggregationcited by 8
  14. 2023Manipulating Transfer Learning for Property Inferencecited by 7
  15. 2024Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attackscited by 7
  16. 2023Samplable Anonymous Aggregation for Private Federated Data Analysiscited by 6
  17. 2023Understanding Deep Gradient Leakage via Inversion Influence Functionscited by 6
  18. 2023SoK: Unintended Interactions among Machine Learning Defenses and Riskscited by 6
  19. 2024Data Reconstruction Attacks and Defenses: A Systematic Evaluationcited by 5
  20. 2024Breaking Secure Aggregation: Label Leakage from Aggregated Gradients in Federated Learningcited by 5
  21. 2025CENSOR: Defense Against Gradient Inversion via Orthogonal Subspace Bayesian Samplingcited by 5
  22. 2025GRAIN: Exact Graph Reconstruction from Gradientscited by 5
  23. 2023Gradient Leakage Defense with Key-Lock Module for Federated Learningcited by 4
  24. 2023Client-side Gradient Inversion Against Federated Learning from Poisoningcited by 4
  25. 2023Maximum Knowledge Orthogonality Reconstruction with Gradients in Federated Learningcited by 4
  26. 2024Provable Privacy Attacks on Trained Shallow Neural Networkscited by 4
  27. 2024Gradient Inversion Attack on Graph Neural Networkscited by 4
  28. 2024FedDP: Privacy-preserving method based on federated learning for histopathology image segmentationcited by 3
  29. 2024Geminio: Language-Guided Gradient Inversion Attacks in Federated Learningcited by 3
  30. 2025Gradient Inversion Attacks on Parameter-Efficient Fine-Tuningcited by 3
  31. 2025Research in Collaborative Learning Does Not Serve Cross-Silo Federated Learning in Practicecited by 3
  32. 2022Perfectly Accurate Membership Inference by a Dishonest Central Server in Federated Learningcited by 2
  33. 2022Combining Stochastic Defenses to Resist Gradient Inversion: An Ablation Studycited by 2
  34. 2023WW-FL: Secure and Private Large-Scale Federated Learningcited by 2
  35. 2023Privacy Preserving Bayesian Federated Learning in Heterogeneous Settingscited by 2
  36. 2024Leak and Learn: An Attacker's Cookbook to Train Using Leaked Data from Federated Learningcited by 2
  37. 2024GI-SMN: Gradient Inversion Attack against Federated Learning without Prior Knowledgecited by 2
  38. 2024Dealing Doubt: Unveiling Threat Models in Gradient Inversion Attacks under Federated Learning, A Survey and Taxonomycited by 2

50 most cited of 72 All 72 (JSON)