Research map / Federated learning and privacy / Privacy attacks and secure learning
Learning to Attack Federated Learning: A Model-based Reinforcement Learning Attack Framework
Henger Li, Xiaolin Sun, Zizhan Zheng
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 |
|---|---|---|---|
| add distribution learning (gradient inversion) to RL attack (ablation) | RL-based attack · MNIST non-iid q=0.3 | global model accuracy: up to 20% -> below 10% | Improved |
| RL attack without any distribution learning vs baselines (ablation) | RL-based attack · MNIST | global model accuracy | Improved |
| fewer attackers: 8 of 100 clients (ablation) | RL-based attack · MNIST | global model accuracy | Improved |
| fewer attackers: 7 of 100 clients (ablation) | RL-based attack · MNIST | global model accuracy | No clear effect |
| longer policy learning (ablation) | RL-based attack · MNIST | simulated global model loss | Improved |
| RL attack vs IPM, LMP, EB baselines under Krum | RL-based attack · MNIST Krum | global model accuracy: drives accuracy to about 10% | Improved |
| vary non-iid degree q (0.1, 0.3, 0.5) (ablation) | RL-based attack · MNIST | attack accuracy variance | Mixed |
Closest papers by meaning
- Defending Against Sophisticated Poisoning Attacks with RL-based Aggregation in Federated Learning (AAAI 2025 · similarity 0.90)
- Model Poisoning Attacks to Federated Learning via Multi-Round Consistency (CVPR 2025 · similarity 0.90)
- FedREDefense: Defending against Model Poisoning Attacks for Federated Learning using Model Update Reconstruction Error (ICML 2024 · similarity 0.88)
- Tracing Back the Malicious Clients in Poisoning Attacks to Federated Learning (NeurIPS 2025 · similarity 0.88)
- FedAA: A Reinforcement Learning Perspective on Adaptive Aggregation for Fair and Robust Federated Learning (AAAI 2025 · similarity 0.86)
- RFLPA: A Robust Federated Learning Framework against Poisoning Attacks with Secure Aggregation (NeurIPS 2024 · similarity 0.86)
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.
