Byzantine Robust Federated Learning Scheme Based on Backdoor Triggers
Source abstract
Abstract
Federated learning is widely used to solve the problem of data decentralization and can provide privacy protection for data owners. However, since multiple participants are required in federated learning, this allows attackers to compromise. Byzantine attacks pose great threats to federated learning. Byzantine attackers upload maliciously created local models to the server to affect the prediction performance and training speed of the global model. To defend against Byzantine attacks, we propose a Byzantine robust federated learning scheme based on backdoor triggers. In our scheme, backdoor triggers are embedded into benign data samples, and then malicious local models can be identified by the server according to its validation dataset. Furthermore, we calculate the adjustment factors of local models according to the parameters of their final layers, which are used to defend against data poisoning-based Byzantine attacks. To further enhance the robustness of our scheme, each local model is weighted and aggregated according to the number of times it is identified as malicious. Relevant experimental data show that our scheme is effective against Byzantine attacks in both independent identically distributed (IID) and non-independent identically distributed (non-IID) scenarios.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Privacy-Preserving Technologies in Data · Cryptography and Data Security · Stochastic Gradient Optimization Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (24)
Show source-provided references
1. J. Konecný, H. B. McMahan, F. X. Yu, P. Richtárik, A. T. Suresh and D. Bacon, “Federated learning: Strategies for improving communication efficiency,” arXiv preprint arXiv:1610.05492, Oct. 2016.
2. J. Konecný, H. B. McMahan, D. Ramage, and P. Richtárik, “Federated optimization: Distributed machine learning for on-device intelligence,” arXiv preprint arXiv:1610.02527, Oct. 2016.
3. Federated split learning for sequential data in satellite-terrestrial integrated networks10.1016/j.inffus.2023.102141 ↗
4. H. B. McMahan, E. Moore, D. Ramage, and B. A. y Arcas, “Federated learning of deep networks using model averaging,” arXiv preprint arXiv:1602.05629, Feb. 2016.
5. B. Kailkhura, S. Brahma, and P. K. Varshney, “Optimal Byzantine attacks on distributed detection in tree-based topologies,” presented at Int. Conf. Comput., Netw. Commun., San Diego, CA, USA, 2013, pp. 227–231.10.1109/ICCNC.2013.6504085 ↗
6. Poisoning attacks on federated learning-based iot intrusion detection system
7. H. Wang et al., “Attack of the tails: Yes, you really can backdoor federated learning,” presented at Adv. Neural Inf. Proces. Syst., 2020, pp. 16070–16084.
8. Coordinated backdoor attacks against federated learning with model-dependent triggers10.1109/MNET.011.2000783 ↗
9. How to backdoor federated learning
10. Z. Zhang et al., “Neurotoxin: Durable backdoors in federated learning,” presented at ICML, Baltimore, MD, USA, 2022, pp. 26429–26446.
11. Byzantine tolerant algorithms for federated learning
12. Differentially private byzantine-robust federated learning10.1109/TPDS.2022.3167434 ↗
13. Byzantine-resilient high-dimensional federated learning10.1109/TIT.2023.3284427 ↗
14. Dynamic defense against Byzantine poisoning attacks in federated learning10.1016/j.future.2022.03.003 ↗
15. Byzantine-resilient federated learning at edge10.1109/TC.2023.3257510 ↗
16. Byzantine-robust federated learning via credibility assessment on non-iid data10.3934/mbe.2022078 ↗
17. Collaborative byzantine resilient federated learning10.1109/JIOT.2023.3266347 ↗
18. Defending Byzantine attacks in ensemble federated learning10.1016/j.future.2023.05.002 ↗
19. Byzantine-resilient secure federated learning10.1109/JSAC.2020.3041404 ↗
20. Gradient-based learning applied to document recognition10.1109/5.726791 ↗
21. Krizhevsky · Technical Report (CIFAR) · 2009 · 1
22. P. Blanchard, E. M. E. Mhamdi, R. Guerraoui, and J. Stainer, “Machine learning with adversaries: Byzantine tolerant gradient descent,” present at Adv. Neural Inf. Proces. Syst., Long Beach, CA, USA, 2017, pp. 119–129.
23. D. Yin, Y. Chen, K. Ramchandran, and P. L. Bartlett, “Byzantine-robust distributed learning: Towards optimal statistical rates,” present at ICML, Stockholm, Sweden, 2018, pp. 5636–5645.
24. Byzantine-robust federated learning through collaborative malicious gradient filtering