非独立同分布联邦学习中基于自适应量化与差分隐私的隐私增强与通信效率优化

Enhanced Privacy and Communication Efficiency in Non-IID Federated Learning with Adaptive Quantization and Differential Privacy

April 25, 2026

作者: Emre Ardıç, Yakup Genç

cs.AI

摘要

联邦学习（FL）是一种分布式机器学习方法，允许多个设备在中央服务器协调下协同训练模型而无需共享底层数据。该方法面临的关键挑战之一在于设备间连接速度与带宽差异导致的通信瓶颈，因此必须缩减训练过程中的数据传输量。此外，训练过程中存在通过模型或梯度分析泄露敏感信息的潜在风险。为同时保障隐私性与通信效率，我们结合差分隐私（DP）与自适应量化方法：采用基于拉普拉斯机制的DP技术保护隐私（该方法在FL研究中相对未被充分探索，且能提供比高斯机制更严格的隐私保障）；提出基于轮次的余弦退火全局比特长度调度器，以及通过数据集熵值分析动态评估客户端贡献度的自适应客户端调度器。我们在CIFAR10、MNIST和医学影像数据集上进行了大规模实验，测试场景涵盖非独立同分布数据、不同客户端数量、比特长度调度策略及隐私预算。结果表明，相较于32位浮点训练，自适应量化方法在MNIST数据集上减少通信总量达52.64%，在CIFAR10上达45.06%，在医学影像数据集上达31%-37%，同时保持具有竞争力的模型精度，并通过差分隐私机制确保稳健的隐私保护。

English

Federated learning (FL) is a distributed machine learning method where multiple devices collaboratively train a model under the management of a central server without sharing underlying data. One of the key challenges of FL is the communication bottleneck caused by variations in connection speed and bandwidth across devices. Therefore, it is essential to reduce the size of transmitted data during training. Additionally, there is a potential risk of exposing sensitive information through the model or gradient analysis during training. To address both privacy and communication efficiency, we combine differential privacy (DP) and adaptive quantization methods. We use Laplacian-based DP to preserve privacy, which is relatively underexplored in FL and offers tighter privacy guarantees than Gaussian-based DP. We propose a simple and efficient global bit-length scheduler using round-based cosine annealing, along with a client-based scheduler that dynamically adapts based on client contribution estimated through dataset entropy analysis. We evaluate our approach through extensive experiments on CIFAR10, MNIST, and medical imaging datasets, using non-IID data distributions across varying client counts, bit-length schedulers, and privacy budgets. The results show that our adaptive quantization methods reduce total communicated data by up to 52.64% for MNIST, 45.06% for CIFAR10, and 31% to 37% for medical imaging datasets compared to 32-bit float training while maintaining competitive model accuracy and ensuring robust privacy through differential privacy.