mlsanityData to 14 Oct 2026

Get Rid Of Your Trail: Remotely Erasing Backdoors in Federated Learning

arXiv 2304.10638 ↗First posted 20 Apr 2023cs.LG cs.CRcited by 13

Not on the map. Why?

Abstract (arXiv)

Federated Learning (FL) enables collaborative deep learning training across multiple participants without exposing sensitive personal data. However, the distributed nature of FL and the unvetted participants' data makes it vulnerable to backdoor attacks. In these attacks, adversaries inject malicious functionality into the centralized model during training, leading to intentional misclassifications for specific adversary-chosen inputs. While previous research has demonstrated successful injections of persistent backdoors in FL, the persistence also poses a challenge, as their existence in the centralized model can prompt the central aggregation server to take preventive measures to penalize the adversaries. Therefore, this paper proposes a methodology that enables adversaries to effectively remove backdoors from the centralized model upon achieving their objectives or upon suspicion of possible detection. The proposed approach extends the concept of machine unlearning and presents strategies to preserve the performance of the centralized model and simultaneously prevent over-unlearning of information unrelated to backdoor patterns, making the adversaries stealthy while removing backdoors. To the best of our knowledge, this is the first work that explores machine unlearning in FL to remove backdoors to the benefit of adversaries. Exhaustive evaluation considering image classification scenarios demonstrates the efficacy of the proposed method in efficient backdoor removal from the centralized model, injected by state-of-the-art attacks across multiple configurations.

Key facts. Cited by 13 papers in the map (data through 14 Oct 2026). Builds on 50 of its 71 references that are in the map. First posted on arXiv 20 Apr 2023.

Builds on

50 of 71 references.

References · 50

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  5. 2016Federated Learning: Strategies for Improving Communication Efficiencycited by 1,475
  6. 2019Machine Unlearningcited by 1,061
  7. 2019Towards Federated Learning at Scale: System Designcited by 832
  8. 2018How To Backdoor Federated Learningcited by 765
  9. 2018Byzantine-Robust Distributed Learning: Towards Optimal Statistical Ratescited by 711
  10. 2017Federated Multi-Task Learningcited by 672
  11. 2020The Future of Digital Health with Federated Learningcited by 633
  12. 2019Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networkscited by 618
38 more
  1. 2018Federated Learning for Mobile Keyboard Predictioncited by 572
  2. 2016Federated Optimization: Distributed Machine Learning for On-Device Intelligencecited by 564
  3. 2019Certified Data Removal from Machine Learning Modelscited by 535
  4. 2021Remember What You Want to Forget: Algorithms for Machine Unlearningcited by 371
  5. 2020Amnesiac Machine Learningcited by 319
  6. 2022A Survey of Machine Unlearningcited by 312
  7. 2020Descent-to-Delete: Gradient-Based Methods for Machine Unlearningcited by 311
  8. 2020Attack of the Tails: Yes, You Really Can Backdoor Federated Learningcited by 309
  9. 2019Can You Really Backdoor Federated Learning?cited by 255
  10. 2021Machine Unlearning of Features and Labelscited by 229
  11. 2020Forgetting Outside the Box: Scrubbing Deep Networks of Information Accessible from Input-Output Observationscited by 180
  12. 2021Federated Learning for Open Bankingcited by 141
  13. 2021Back to the Drawing Board: A Critical Evaluation of Poisoning Attacks on Production Federated Learningcited by 118
  14. 2022The Right to be Forgotten in Federated Learning: An Efficient Realization with Rapid Retrainingcited by 118
  15. 2021Federated Unlearning via Class-Discriminative Pruningcited by 115
  16. 2022Neurotoxin: Durable Backdoors in Federated Learningcited by 98
  17. 2022Federated Unlearning: How to Efficiently Erase a Client in FL?cited by 95
  18. 2020Defending against Backdoors in Federated Learning with Robust Learning Ratecited by 82
  19. 2022DeepSight: Mitigating Backdoor Attacks in Federated Learning Through Deep Model Inspectioncited by 81
  20. 2022Federated Unlearning with Knowledge Distillationcited by 73
  21. 2022Backdoor Defense with Machine Unlearningcited by 65
  22. 2020BaFFLe: Backdoor detection via Feedback-based Federated Learningcited by 53
  23. 2021Hard to Forget: Poisoning Attacks on Certified Machine Unlearningcited by 52
  24. 2021SparseFed: Mitigating Model Poisoning Attacks in Federated Learning with Sparsificationcited by 46
  25. 2022VeriFi: Towards Verifiable Federated Unlearningcited by 46
  26. 2022Federated Unlearning for On-Device Recommendationcited by 43
  27. 2020Mitigating Backdoor Attacks in Federated Learningcited by 39
  28. 2022FLIP: A Provable Defense Framework for Backdoor Mitigation in Federated Learningcited by 37
  29. 2023On the Vulnerability of Backdoor Defenses for Federated Learningcited by 27
  30. 2023BayBFed: Bayesian Backdoor Defense for Federated Learningcited by 23
  31. 2021Forget-SVGD: Particle-Based Bayesian Federated Unlearningcited by 16
  32. 2022Vertical Semi-Federated Learning for Efficient Online Advertisingcited by 12
  33. 2020Dynamic backdoor attacks against federated learningcited by 9
  34. 2021Bayesian Variational Federated Learning and Unlearning in Decentralized Networkscited by 9
  35. 2022Thinking Two Moves Ahead: Anticipating Other Users Improves Backdoor Attacks in Federated Learningcited by 7
  36. 2022PerDoor: Persistent Non-Uniform Backdoors in Federated Learning using Adversarial Perturbationscited by 5
  37. 2022Backdoor Defense in Federated Learning Using Differential Testing and Outlier Detectioncited by 4
  38. 2022A Knowledge Distillation-Based Backdoor Attack in Federated Learningcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 8; 2025: 3; 2026: 1.

13 papers.

Sort:

By area:

  1. 2023A Survey on Federated Unlearning: Challenges, Methods, and Future Directionscited by 63
  2. 2024Federated Unlearning: A Survey on Methods, Design Guidelines, and Evaluation Metricscited by 27
  3. 2024Ferrari: Federated Feature Unlearning via Optimizing Feature Sensitivitycited by 11
  4. 2024Towards Privacy-Guaranteed Label Unlearning in Vertical Federated Learning: Few-Shot Forgetting without Disclosurecited by 5
  5. 2024Identify Backdoored Model in Federated Learning via Individual Unlearningcited by 4
  6. 2024FedQUIT: On-Device Federated Unlearning via a Quasi-Competent Virtual Teachercited by 2
  7. 2024Mind the Cost of Scaffold! Benign Clients May Even Become Accomplices of Backdoor Attackcited by 1
  8. 2025DROP: Poison Dilution via Knowledge Distillation for Federated Learningcited by 1
  9. 2025Fully Decentralized Certified Unlearningcited by 1
  10. 2026Jellyfish: Zero-Shot Federated Unlearning Scheme with Knowledge Disentanglementcited by 1
  11. 2024A Survey on Federated Unlearning: Challenges and Opportunitiescited by 0
  12. 2024Upcycling Noise for Federated Unlearningcited by 0
1 more
  1. 2025DRAUN: An Algorithm-Agnostic Data Reconstruction Attack on Federated Unlearning Systemscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025BadFU: Backdoor Federated Learning through Adversarial Machine Unlearning
  2. 2022Towards a Defense Against Federated Backdoor Attacks Under Continuous Training
  3. 2024Client-Side Patching against Backdoor Attacks in Federated Learning
  4. 2024Concealing Backdoor Model Updates in Federated Learning by Trigger-Optimized Data Poisoning
  5. 2024Vertical Federated Unlearning via Backdoor Certification
  6. 2024Non-Cooperative Backdoor Attacks in Federated Learning: A New Threat Landscape
  7. 2025Enhancing the Effectiveness and Durability of Backdoor Attacks in Federated Learning through Maximizing Task Distinction
  8. 2024FLAIN: Mitigating Backdoor Attacks in Federated Learning via Flipping Weight Updates of Low-Activation Input Neurons