Research map / Adversarial robustness and security / Backdoor and poisoning
FedGame: A Game-Theoretic Defense against Backdoor Attacks in Federated Learning
Jinyuan Jia, Zhuowen Yuan, Dinuka Sahabandu, Luyao Niu, Arezoo Rajabi, Bhaskar Ramasubramanian, Bo Li, Radha Poovendran
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| replace Neural Cleanse with FeatureRE for trigger reverse engineering (ablation) | FedGame · MNIST Scaling attack | ASR (%): 10.36 | No clear effect |
| use low-quality reverse-engineered trigger (20 iterations instead of 100) (ablation) | FedGame · MNIST Scaling attack | ASR (%): 13.23 | No clear effect |
| increase total number of clients from 10 to 100 (ablation) | FedGame · MNIST | ASR (%): 9.72 -> 9.89 | No clear effect |
| increase total number of clients from 10 to 100 (ablation) | FedGame · CIFAR10 | ASR (%): 8.92 -> 10.82 | No clear effect |
| reduce server clean data ratio to 0.01 (ablation) | FedGame · MNIST | ASR (%): 9.71 | No clear effect |
| reduce server clean data ratio to 0.02 (ablation) | FedGame · MNIST | ASR (%): 12.38 | No clear effect |
| reduce server clean data ratio to 0.05 (ablation) | FedGame · MNIST | ASR (%): 9.75 | No clear effect |
| attacker optimizes trigger pattern (ablation) | FedGame · MNIST | ASR (%): 12.43 | No clear effect |
| attacker replaces local data with backdoored samples (ablation) | FedGame · MNIST | ASR (%): 9.71 | No clear effect |
| attacker replaces local data with backdoored samples (ablation) | FedGame · CIFAR10 | ASR (%): 10.69 | No clear effect |
| increase compromised clients to 80%: FedGame versus FLTrust | FedGame vs FLTrust · MNIST | ASR (%): 99.95 -> 9.84 | Improved |
| static attacker not optimizing backdoor fraction (ablation) | FedGame · default setting | ASR (%): 9.75 | No clear effect |
Closest papers by meaning
- Defending against Backdoors in Federated Learning with Robust Learning Rate (AAAI 2021 · similarity 0.92)
- Resisting Backdoor Attacks in Federated Learning via Bidirectional Elections and Individual Perspective (AAAI 2024 · similarity 0.90)
- Beyond Traditional Threats: A Persistent Backdoor Attack on Federated Learning (AAAI 2024 · similarity 0.89)
- Backdoor Federated Learning by Poisoning Backdoor-Critical Layers (ICLR 2024 · similarity 0.88)
- Infighting in the Dark: Multi-Label Backdoor Attack in Federated Learning (CVPR 2025 · similarity 0.87)
- MARS: A Malignity-Aware Backdoor Defense in Federated Learning (NeurIPS 2025 · similarity 0.87)
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.
