通过惩罚梯度范数，提升私有联邦学习的全局平坦极小值，改善泛化性能。
DP-FedPGN: Finding Global Flat Minima for Differentially Private Federated Learning via Penalizing Gradient Norm
- 引入全局梯度范数惩罚，寻找全局平坦极小值而非局部平坦解。
- 在6个视觉与自然语言任务上超越现有最优方法，显著提升性能。
- 适用于数据异构场景，兼具严格隐私保障与快速收敛能力。
为防止联邦学习中的推断攻击并降低敏感信息泄露，客户端级差分隐私联邦学习（CL-DPFL）被广泛应用。然而，现有方法通常导致更尖锐的损失曲面，损害模型泛化能力。尽管当前主流方法采用锐度感知最小化（SAM）寻找局部平坦极小值，但其局部平坦性未必反映全局平坦性。为此，本文提出新型CL-DPFL算法DP-FedPGN，通过在本地损失中引入全局梯度范数惩罚，以寻找全局平坦极小值。该方法不仅获得更平坦的全局最小值，还降低了本地更新的梯度范数，从而进一步减少梯度裁剪误差。理论上分析了DP-FedPGN如何缓解差分隐私带来的性能下降，并证明其可消除数据异质性影响，实现快速收敛。使用Rényi差分隐私提供严格隐私保障，并对本地更新进行灵敏度分析。在ResNet和Transformer模型上进行测试，在六个视觉与自然语言处理任务中均显著优于现有最先进方法。
原文摘要 · Abstract (English)
To prevent inference attacks in Federated Learning (FL) and reduce the leakage of sensitive information, Client-level Differentially Private Federated Learning (CL-DPFL) is widely used. However, current CL-DPFL methods usually result in sharper loss landscapes, which leads to a decrease in model generalization after differential privacy protection. By using Sharpness Aware Minimization (SAM), the current popular federated learning methods are to find a local flat minimum value to alleviate this problem. However, the local flatness may not reflect the global flatness in CL-DPFL. Therefore, to address this issue and seek global flat minima of models, we propose a new CL-DPFL algorithm, DP-FedPGN, in which we introduce a global gradient norm penalty to the local loss to find the global flat minimum. Moreover, by using our global gradient norm penalty, we not only find a flatter global minimum but also reduce the locally updated norm, which means that we further reduce the error of gradient clipping. From a theoretical perspective, we analyze how DP-FedPGN mitigates the performance degradation caused by DP. Meanwhile, the proposed DP-FedPGN algorithm eliminates the impact of data heterogeneity and achieves fast convergence. We also use Rényi DP to provide strict privacy guarantees and provide sensitivity analysis for local updates. Finally, we conduct effectiveness tests on both ResNet and Transformer models, and achieve significant improvements in six visual and natural language processing tasks compared to existing state-of-the-art algorithms. The code is available at https://github.com/junkangLiu0/DP-FedPGN
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。