mlsanityData to 14 Oct 2026

Federated Unlearning via Class-Discriminative Pruning

arXiv 2110.11794 ↗First posted 22 Oct 2021cs.CV cs.CR cs.DCcited by 115

On the map: LLM safety and security → Federated unlearning and poisoning

Abstract (arXiv)

We explore the problem of selectively forgetting categories from trained CNN classification models in the federated learning (FL). Given that the data used for training cannot be accessed globally in FL, our insights probe deep into the internal influence of each channel. Through the visualization of feature maps activated by different channels, we observe that different channels have a varying contribution to different categories in image classification. Inspired by this, we propose a method for scrubbing the model clean of information about particular categories. The method does not require retraining from scratch, nor global access to the data used for training. Instead, we introduce the concept of Term Frequency Inverse Document Frequency (TF-IDF) to quantize the class discrimination of channels. Channels with high TF-IDF scores have more discrimination on the target categories and thus need to be pruned to unlearn. The channel pruning is followed by a fine-tuning process to recover the performance of the pruned model. Evaluated on CIFAR10 dataset, our method accelerates the speed of unlearning by 8.9x for the ResNet model, and 7.9x for the VGG model under no degradation in accuracy, compared to retraining from scratch. For CIFAR100 dataset, the speedups are 9.9x and 8.4x, respectively. We envision this work as a complementary block for FL towards compliance with legal and ethical criteria.

Key facts. Cited by 115 papers in the map (data through 14 Oct 2026). Placed in Federated unlearning and poisoning, one of 12 directions in LLM safety and security. Builds on 28 of its 37 references that are in the map. First posted on arXiv 22 Oct 2021.

Builds on

28 of 37 references.

References · 28

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  3. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  4. 2019Advances and Open Problems in Federated Learningcited by 3,046
  5. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
  6. 2017Understanding Black-box Predictions via Influence Functionscited by 1,741
  7. 2016Federated Learning: Strategies for Improving Communication Efficiencycited by 1,475
  8. 2018Federated Learning with Non-IID Datacited by 1,108
  9. 2019Machine Unlearningcited by 1,061
  10. 2017Channel Pruning for Accelerating Very Deep Neural Networkscited by 702
  11. 2019Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networkscited by 618
  12. 2019Certified Data Removal from Machine Learning Modelscited by 535
16 more
  1. 2019Local Model Poisoning Attacks to Byzantine-Robust Federated Learningcited by 493
  2. 2019Making AI Forget You: Data Deletion in Machine Learningcited by 465
  3. 2014New insights and perspectives on the natural gradient methodcited by 443
  4. 2019Federated Machine Learning: Concept and Applicationscited by 412
  5. 2015Understanding Neural Networks Through Deep Visualizationcited by 334
  6. 2020Amnesiac Machine Learningcited by 319
  7. 2020Descent-to-Delete: Gradient-Based Methods for Machine Unlearningcited by 311
  8. 2020Approximate Data Deletion from Machine Learning Modelscited by 255
  9. 2020Forgetting Outside the Box: Scrubbing Deep Networks of Information Accessible from Input-Output Observationscited by 180
  10. 2018Discrimination-aware Channel Pruning for Deep Neural Networkscited by 171
  11. 2020Mixed-Privacy Forgetting in Deep Networkscited by 163
  12. 2020DeltaGrad: Rapid retraining of machine learning modelscited by 149
  13. 2020Machine Unlearning for Random Forestscited by 131
  14. 2020Variational Bayesian Unlearningcited by 117
  15. 2020Channel Pruning via Automatic Structure Searchcited by 85
  16. 2020PrIU: A Provenance-Based Approach for Incrementally Updating Regression Modelscited by 16

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 9; 2023: 19; 2024: 45; 2025: 27; 2026: 15.

115 papers.

Sort:

By area:

  1. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generationcited by 386
  2. 2022A Survey of Machine Unlearningcited by 312
  3. 2024Rethinking Machine Unlearning for Large Language Modelscited by 257
  4. 2023Model Sparsity Can Simplify Machine Unlearningcited by 214
  5. 2023To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Nowcited by 210
  6. 2022Can Bad Teaching Induce Forgetting? Unlearning in Deep Networks using an Incompetent Teachercited by 194
  7. 2023Boundary Unlearningcited by 134
  8. 2024SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearningcited by 104
  9. 2022Federated Unlearning: How to Efficiently Erase a Client in FL?cited by 95
  10. 2023Machine Unlearning: Solutions and Challengescited by 93
  11. 2024What makes unlearning hard and what to do about itcited by 89
  12. 2023A Survey on Federated Unlearning: Challenges, Methods, and Future Directionscited by 63
38 more
  1. 2024Scissorhands: Scrub Data Influence via Connection Sensitivity in Networkscited by 52
  2. 2023Exploring the Landscape of Machine Unlearning: A Comprehensive Survey and Taxonomycited by 51
  3. 2024Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearningcited by 44
  4. 2023GNNDelete: A General Strategy for Unlearning in Graph Neural Networkscited by 43
  5. 2022Evaluating Machine Unlearning via Epistemic Uncertaintycited by 39
  6. 2024Threats, Attacks, and Defenses in Machine Unlearning: A Surveycited by 39
  7. 2024Revisiting Who's Harry Potter: Towards Targeted Unlearning from a Causal Intervention Perspectivecited by 36
  8. 2024Machine Unlearning: A Comprehensive Surveycited by 35
  9. 2023Backdoor Attacks and Defenses in Federated Learning: Survey, Challenges and Future Research Directionscited by 33
  10. 2022Deep Regression Unlearningcited by 32
  11. 2023SAFE: Machine Unlearning With Shard Graphscited by 28
  12. 2024Federated Unlearning: A Survey on Methods, Design Guidelines, and Evaluation Metricscited by 27
  13. 2024When Machine Unlearning Meets Retrieval-Augmented Generation (RAG): Keep Secret or Forget Knowledge?cited by 24
  14. 2024Learning to Refuse: Towards Mitigating Privacy Risks in LLMscited by 23
  15. 2022Machine Unlearning of Federated Clusterscited by 22
  16. 2025Ten Challenging Problems in Federated Foundation Modelscited by 20
  17. 2023Machine Unlearning: A Surveycited by 19
  18. 2025Learning to Unlearn while Retaining: Combating Gradient Conflicts in Machine Unlearningcited by 19
  19. 2022SIFU: Sequential Informed Federated Unlearning for Efficient and Provable Client Unlearning in Federated Optimizationcited by 18
  20. 2024MEOW: MEMOry Supervised LLM Unlearning Via Inverted Factscited by 18
  21. 2022Efficient Attribute Unlearning: Towards Selective Removal of Input Attributes from Feature Representationscited by 17
  22. 2023Subspace based Federated Unlearningcited by 17
  23. 2024MU-Bench: A Multitask Multimodal Benchmark for Machine Unlearningcited by 17
  24. 2024MUNBa: Machine Unlearning via Nash Bargainingcited by 17
  25. 2025Unlearning through Knowledge Overwriting: Reversible Federated Unlearning via Selective Sparse Adaptercited by 16
  26. 2024A Survey on Machine Unlearning: Techniques and New Emerged Privacy Riskscited by 15
  27. 2024Unlearning during Learning: An Efficient Federated Machine Unlearning Methodcited by 14
  28. 2023Get Rid Of Your Trail: Remotely Erasing Backdoors in Federated Learningcited by 13
  29. 2023QuickDrop: Efficient Federated Unlearning by Integrated Dataset Distillationcited by 11
  30. 2024Ferrari: Federated Feature Unlearning via Optimizing Feature Sensitivitycited by 11
  31. 2024Don't Forget Too Much: Towards Machine Unlearning on Feature Levelcited by 11
  32. 2024Machine Unlearning for Traditional Models and Large Language Models: A Short Surveycited by 10
  33. 2025UIPE: Enhancing LLM Unlearning by Removing Knowledge Related to Forgetting Targetscited by 10
  34. 2024Label Smoothing Improves Machine Unlearningcited by 9
  35. 2025Data Duplication: A Novel Multi-Purpose Attack Paradigm in Machine Unlearningcited by 9
  36. 2025ESC: Erasing Space Concept for Knowledge Deletioncited by 9
  37. 2023Unlearning with Fisher Maskingcited by 8
  38. 2024Fast-FedUL: A Training-Free Federated Unlearning with Provable Skew Resiliencecited by 8

50 most cited of 115 All 115 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Forgetting to Witness: Efficient Federated Unlearning and Its Visible Evaluation
  2. 2023Deep Unlearning: Fast and Efficient Gradient-free Approach to Class Forgetting
  3. 2025REMISVFU: Vertical Federated Unlearning via Representation Misdirection for Intermediate Output Feature
  4. 2023Multi-Class Unlearning for Image Classification via Weight Filtering
  5. 2026Jellyfish: Zero-Shot Federated Unlearning Scheme with Knowledge Disentanglement
  6. 2026f-FUM: Federated Unlearning via min--max and f-divergence
  7. 2025Certifying the Right to Be Forgotten: Primal-Dual Optimization for Sample and Label Unlearning in Vertical Federated Learning
  8. 2024Goldfish: An Efficient Federated Unlearning Framework