提出统一框架，同时实现隐私保护、抗干扰和通信高效。
Beyond Trade-offs: A Unified Framework for Privacy, Robustness, and Communication Efficiency in Federated Learning
- 用鲁棒兼容压缩降低双向通信开销。
- 在CIFAR-10等数据集上保持高鲁棒性和隐私性。
- 适合对安全与效率并重的分布式学习场景。
我们提出 Fed-DPRoC，一种新型联邦学习框架，可同时提供差分隐私（DP）、拜占庭鲁棒性和通信效率。核心是鲁棒兼容压缩，可在不损害聚合鲁棒性的前提下减少双向通信开销。我们构建了 RobAJoL 实例，结合基于 Johnson-Lindenstrauss（JL）变换的压缩机制与鲁棒平均方法以保障鲁棒性。理论分析证明了 JL 变换与鲁棒平均的兼容性，确保 RobAJoL 在满足差分隐私的同时，显著降低通信开销。在 CIFAR-10、Fashion MNIST 和 FEMNIST 上的仿真结果验证了理论结论。与加入差分隐私的先进通信高效鲁棒方案相比，RobAJoL 在不同拜占庭攻击下均展现出更优的鲁棒性与模型性能。
原文摘要 · Abstract (English)
We propose Fed-DPRoC, a novel federated learning framework designed to jointly provide differential privacy (DP), Byzantine robustness, and communication efficiency. Central to our approach is the concept of robust-compatible compression, which allows reducing the bi-directional communication overhead without undermining the robustness of the aggregation. We instantiate our framework as RobAJoL, which integrates the Johnson-Lindenstrauss (JL)-based compression mechanism with robust averaging for robustness. Our theoretical analysis establishes the compatibility of JL transform with robust averaging, ensuring that RobAJoL maintains robustness guarantees, satisfies DP, and substantially reduces communication overhead. We further present simulation results on CIFAR-10, Fashion MNIST, and FEMNIST, validating our theoretical claims. We compare RobAJoL with a state-of-the-art communication-efficient and robust FL scheme augmented with DP for a fair comparison, demonstrating that RobAJoL outperforms existing methods in terms of robustness and utility under different Byzantine attacks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。