Research map / Adversarial robustness and security / Certified and empirical robustness
Delving into the Adversarial Robustness of Federated Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add DBFAT local re-weighting (ablation) | DBFAT · CIFAR10 non-IID | Clean accuracy: 48.44 -> 52.16 | Helped |
| Add DBFAT local re-weighting (ablation) | DBFAT · CIFAR10 non-IID | Average robust accuracy: 25.89 -> 27.80 | Helped |
| Add DBFAT global regularization (ablation) | DBFAT · CIFAR10 non-IID | Clean accuracy: 51.04 -> 52.16 | Helped |
| Add DBFAT global regularization (ablation) | DBFAT · CIFAR10 non-IID | Average robust accuracy: 26.84 -> 27.80 | Helped |
| Add DBFAT local re-weighting (ablation) | DBFAT · FMNIST non-IID | Clean accuracy: 72.35 -> 75.89 | Helped |
| Add DBFAT local re-weighting (ablation) | DBFAT · FMNIST non-IID | Average robust accuracy: 56.34 -> 59.63 | Helped |
| Add DBFAT global regularization (ablation) | DBFAT · FMNIST non-IID | Clean accuracy: 73.96 -> 75.89 | Helped |
| Add DBFAT global regularization (ablation) | DBFAT · FMNIST non-IID | Average robust accuracy: 58.23 -> 59.63 | Helped |
| Set regularization beta to 1.5 (ablation) | DBFAT · MNIST | Average robust accuracy: 26.64 (beta 4); 34.24 (beta 2); 53.22 (beta 1.5); 47.35 (beta 1); 44.09 (beta 0.5) | Mixed |
| Set regularization beta to 2 (ablation) | DBFAT · FMNIST | Average robust accuracy: 37.36 (beta 4); 47.83 (beta 2); 44.08 (beta 1.5); 42.33 (beta 1); 45.28 (beta 0.5) | Mixed |
| Use DBFAT instead of AVMixup | FedAvg · CIFAR10 non-IID | PGD-20 accuracy: 24.25 -> 27.02 | Helped |
| Use DBFAT instead of AVMixup | FedAvg · ImageNet-12 non-IID | AA accuracy: 21.92 -> 22.08 | Helped |
Closest papers by meaning
- Federated Robustness Propagation: Sharing Adversarial Robustness in Heterogeneous Federated Learning (AAAI 2023 · similarity 0.93)
- Byzantine-Robust Federated Learning: Impact of Client Subsampling and Local Updates (ICML 2024 · similarity 0.87)
- Towards the Robustness of Differentially Private Federated Learning (AAAI 2024 · similarity 0.86)
- How Does the Smoothness Approximation Method Facilitate Generalization for Federated Adversarial Learning? (AAAI 2025 · similarity 0.85)
- Constructing Adversarial Examples for Vertical Federated Learning: Optimal Client Corruption through Multi-Armed Bandit (ICLR 2024 · similarity 0.85)
- Fine-Tuning Personalization in Federated Learning to Mitigate Adversarial Clients (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.