Defending Federated Learning System from Poisoning Attacks via Efficient Unlearning
Source abstract
Abstract
Large-scale neural networks-based federated learning (FL) has gained public recognition for its effective capabilities in distributed training. Nonetheless, the open system architecture inherent to federated learning systems ... | Find, read and cite all the research you need on Tech Science Press
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Cryptography and Data Security · Cryptography and Residue Arithmetic · Machine Learning and Algorithms
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (31)
Show source-provided references
1. McMahan · Proceedings of the 20th International Conference on Artificial Intelligence and Statistics · 2017 · 1273
2. Federated learning for wireless communications: motivation, opportunities, and challenges10.1109/MCOM.001.1900461 ↗
3. Blanchard · Advances in neural information processing systems · 2017 · 30
4. Comprehensive privacy analysis of deep learning: passive and active white-box inference attacks against centralized and federated learning
5. Poisoning and evasion attacks against deep learning algorithms in autonomous vehicles10.1109/TVT.2020.2977378 ↗
6. Byzantine-robust aggregation with gradient difference compression and stochastic variance reduction for federated learning
7. Yin · Proceedings of the 35th International Conference on Machine Learning · 2018 · 5650
8. Byzantine-resilient secure federated learning10.1109/JSAC.2020.3041404 ↗
9. Defense against local model poisoning attacks to byzantine-robust federated learning10.1007/s11704-021-1067-4 ↗
10. Byzantine-robust aggregation in federated learning empowered industrial IoT10.1109/TII.2021.3128164 ↗
11. Federated unlearning: guarantee the right of clients to forget10.1109/MNET.001.2200198 ↗
12. Wang · WWW ’22: The ACM Web Conference 2022; 2022 Apr 25–29; Lyon, France · 2022 · 622
13. Bayesian variational federated learning and unlearning in decentralized networks
14. Machine unlearning: solutions and challenges10.1109/TETCI.2024.3379240 ↗
15. Understanding distributed poisoning attack in federated learning
16. FLTrust: byzantine-robust federated learning via trust bootstrapping
17. On the byzantine robustness of clustered federated learning
18. Zhai K, Ren Q, Wang J, Yan C. Byzantine-robust federated learning via credibility assessment on non-IID data. arXiv:2109.02396. 2021.10.3934/mbe.2022078 ↗
19. Low dimensional secure federated learning framework against poisoning attacks10.1016/j.future.2024.04.017 ↗
20. VPPFL: a verifiable privacy-preserving federated learning scheme against poisoning attacks10.1016/j.cose.2023.103562 ↗
21. Hu T. Dense in dense: training segmentation from scratch. In: Jawahar CV, Li H, Mori G, Schindler K, editors. Computer vision–ACCV 2018. Cham: Springer International Publishing; 2019. p. 454–70.10.1007/978-3-030-20876-9_29 ↗
22. Towards making systems forget with machine unlearning
23. Ginart · Advances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems 2019, NeurIPS 2019; 2019 Dec 8–14 · 2019 · 3513
24. FedEraser: enabling efficient client-level data removal from federated learning models
25. FedRecover: recovering from poisoning attacks in federated learning using historical information
26. FedRecovery: differentially private machine unlearning for federated learning frameworks10.1109/TIFS.2023.3297905 ↗
27. Toward efficient and robust federated unlearning in IoT networks10.1109/JIOT.2024.3378329 ↗
28. FL-defender: combating targeted attacks in federated learning10.1016/j.knosys.2022.110178 ↗
29. Principal component analysis10.1016/0169-7439(87)80084-9 ↗
30. LeCun Y. The MNIST database of handwritten digits; 1998 [cited 2025 Feb 13]. Available from: http://yann.lecun.com/exdb/mnist/.
31. Xiao H, Rasul K, Vollgraf R. Fashion-MNIST: a novel image dataset for benchmarking machine learning algorithms. arXiv:1708.07747. 2017.