mlsanityData to 7 Oct 2026

NoT: Federated Unlearning via Weight Negation

arXiv 2503.05657 ↗First posted 7 Mar 2025cs.LG cs.CVcited by 15

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

Abstract (arXiv)

Federated unlearning (FU) aims to remove a participant's data contributions from a trained federated learning (FL) model, ensuring privacy and regulatory compliance. Traditional FU methods often depend on auxiliary storage on either the client or server side or require direct access to the data targeted for removal-a dependency that may not be feasible if the data is no longer available. To overcome these limitations, we propose NoT, a novel and efficient FU algorithm based on weight negation (multiplying by -1), which circumvents the need for additional storage and access to the target data. We argue that effective and efficient unlearning can be achieved by perturbing model parameters away from the set of optimal parameters, yet being well-positioned for quick re-optimization. This technique, though seemingly contradictory, is theoretically grounded: we prove that the weight negation perturbation effectively disrupts inter-layer co-adaptation, inducing unlearning while preserving an approximate optimality property, thereby enabling rapid recovery. Experimental results across three datasets and three model architectures demonstrate that NoT significantly outperforms existing baselines in unlearning efficacy as well as in communication and computational efficiency.

Key facts. Cited by 15 papers in the map (data through 7 Oct 2026). Placed in Federated unlearning and poisoning, one of 12 directions in LLM safety and security. Builds on 41 of its 70 references that are in the map. First posted on arXiv 7 Mar 2025.

Builds on

41 of 70 references.

Closest ancestors · 7

Same direction, newest first

  1. 2024Federated Unlearning with Gradient Descent and Conflict Mitigation28 Dec
  2. 2024Enable the Right to be Forgotten with Federated Client Unlearning in Medical Imaging2 Jul
  3. 2024Federated Unlearning: A Survey on Methods, Design Guidelines, and Evaluation Metrics10 Jan
  4. 2023A Survey on Federated Unlearning: Challenges, Methods, and Future Directions31 Oct
  5. 2022Federated Unlearning: How to Efficiently Erase a Client in FL?12 Jul
  6. 2022VeriFi: Towards Verifiable Federated Unlearning25 May
  7. 2022Federated Unlearning with Knowledge Distillation24 Jan

Other references · 34

  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. 2019Similarity of Neural Network Representations Revisitedcited by 1,695
  5. 2014How transferable are features in deep neural networks?cited by 1,216
  6. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,052
  7. 2013Exact solutions to the nonlinear dynamics of learning in deep linear neural networkscited by 748
  8. 2019Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networkscited by 599
  9. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generationcited by 376
  10. 2021Membership Inference Attacks on Machine Learning: A Surveycited by 306
  11. 2017Stochastic Gradient Descent as Approximate Bayesian Inferencecited by 298
  12. 2021Unrolling SGD: Understanding Factors Influencing Machine Unlearningcited by 274
22 more
  1. 2020ReZero is All You Need: Fast Convergence at Large Depthcited by 260
  2. 2021Federated Learning for Internet of Things: A Comprehensive Surveycited by 247
  3. 2021Fast Yet Effective Machine Unlearningcited by 209
  4. 2023Model Sparsity Can Simplify Machine Unlearningcited by 206
  5. 2022Can Bad Teaching Induce Forgetting? Unlearning in Deep Networks using an Incompetent Teachercited by 189
  6. 2023Fast Machine Unlearning Without Retraining Through Selective Synaptic Dampeningcited by 177
  7. 2018Dynamical Isometry and a Mean Field Theory of CNNs: How to Train 10,000-Layer Vanilla Convolutional Neural Networkscited by 176
  8. 2020Forgetting Outside the Box: Scrubbing Deep Networks of Information Accessible from Input-Output Observationscited by 171
  9. 2022Signal Propagation in Transformers: Theoretical Perspectives and the Role of Rank Collapsecited by 136
  10. 2017Resurrecting the sigmoid in deep learning through dynamical isometry: theory and practicecited by 132
  11. 2018Universal Statistics of Fisher Information in Deep Neural Networks: Mean Field Approachcited by 89
  12. 2023Machine Unlearning: Solutions and Challengescited by 85
  13. 2021Federated Learning for Internet of Things: Applications, Challenges, and Opportunitiescited by 49
  14. 2018Dynamical Isometry and a Mean Field Theory of RNNs: Gating Enables Signal Propagation in Recurrent Neural Networkscited by 44
  15. 2024Threats, Attacks, and Defenses in Machine Unlearning: A Surveycited by 34
  16. 2024Advances and Open Challenges in Federated Foundation Modelscited by 33
  17. 2018Dynamical Isometry is Achieved in Residual Networks in a Universal Way for any Activation Functioncited by 22
  18. 2023Data and Model Poisoning Backdoor Attacks on Wireless Federated Learning, and the Defense Mechanisms: A Comprehensive Surveycited by 17
  19. 2018Approximate Fisher Information Matrix to Characterise the Training of Deep Neural Networkscited by 12
  20. 2020STNet: Scale Tree Network with Multi-level Auxiliator for Crowd Countingcited by 11
  21. 2019Breaking Inter-Layer Co-Adaptation by Classifier Anonymizationcited by 4
  22. 2024DFML: Decentralized Federated Mutual Learningcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 4; 2026: 11.

15 papers.

Sort:

  1. 2025The Right to be Forgotten in Pruning: Unveil Machine Unlearning on Sparse Modelscited by 5
  2. 2026EASE: Federated Multimodal Unlearning via Entanglement-Aware Anchor Closurecited by 5
  3. 2025CoUn: Empowering Machine Unlearning via Contrastive Learningcited by 4
  4. 2026Certified Unlearning in Decentralized Federated Learningcited by 2
  5. 2026FedCARE: Federated Unlearning with Conflict-Aware Projection and Relearning-Resistant Recoverycited by 2
  6. 2025Federated Unlearning Made Practical: Seamless Integration via Negated Pseudo-Gradientscited by 1
  7. 2025FedUP: Efficient Pruning-based Federated Unlearning for Model Poisoning Attackscited by 1
  8. 2026Lethe: Principled Dual-Stream Update for Persistent Knowledge Erasure in Federated Unlearningcited by 0
  9. 2026f-FUM: Federated Unlearning via min--max and f-divergencecited by 0
  10. 2026Computation and Communication Efficient Federated Unlearning via On-server Gradient Conflict Mitigation and Expressioncited by 0
  11. 2026SoK: Unlearnability and Unlearning for Model Dememorizationcited by 0
  12. 2026Rethinking Federated Unlearning via the Lens of Memorizationcited by 0
3 more
  1. 2026pFedUL: Layer-Aware Federated Unlearning for Personalized Federated Learningcited by 0
  2. 2026IREU: Identity-Related Encoder-Only Unlearning for Customized Portrait Generationcited by 0
  3. 2026Lethe: How Hard Is It to Forget? A Benchmark for Federated Unlearning in Medical Imagingcited by 0