Robust Federated Learning under Adversarial Attacks via Loss-Based Client Clustering
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917444360077312 |
|---|---|
| author | Kritharakis, Emmanouil Jakovetic, Dusan Makris, Antonios Tserpes, Konstantinos |
| author_facet | Kritharakis, Emmanouil Jakovetic, Dusan Makris, Antonios Tserpes, Konstantinos |
| contents | Federated Learning (FL) enables collaborative model training across multiple clients without sharing private data. We consider FL scenarios wherein FL clients are subject to adversarial (Byzantine) attacks, while the FL server is trusted (honest) and has a trustworthy side dataset. This may correspond to, e.g., cases where the server possesses trusted data prior to federation, or to the presence of a trusted client that temporarily assumes the server role. Our approach requires only two honest participants, i.e., the server and one client, to function effectively, without prior knowledge of the number of malicious clients. Theoretical analysis demonstrates bounded optimality gaps even under strong Byzantine attacks. Experimental results show that our algorithm significantly outperforms standard and robust FL baselines such as Mean, Trimmed Mean, Median, Krum, and Multi-Krum under various attack strategies including label flipping, sign flipping, and Gaussian noise addition across MNIST, FMNIST, and CIFAR-10 benchmarks using the Flower framework. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_12672 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Robust Federated Learning under Adversarial Attacks via Loss-Based Client Clustering Kritharakis, Emmanouil Jakovetic, Dusan Makris, Antonios Tserpes, Konstantinos Machine Learning Artificial Intelligence Federated Learning (FL) enables collaborative model training across multiple clients without sharing private data. We consider FL scenarios wherein FL clients are subject to adversarial (Byzantine) attacks, while the FL server is trusted (honest) and has a trustworthy side dataset. This may correspond to, e.g., cases where the server possesses trusted data prior to federation, or to the presence of a trusted client that temporarily assumes the server role. Our approach requires only two honest participants, i.e., the server and one client, to function effectively, without prior knowledge of the number of malicious clients. Theoretical analysis demonstrates bounded optimality gaps even under strong Byzantine attacks. Experimental results show that our algorithm significantly outperforms standard and robust FL baselines such as Mean, Trimmed Mean, Median, Krum, and Multi-Krum under various attack strategies including label flipping, sign flipping, and Gaussian noise addition across MNIST, FMNIST, and CIFAR-10 benchmarks using the Flower framework. |
| title | Robust Federated Learning under Adversarial Attacks via Loss-Based Client Clustering |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2508.12672 |