发表机构
School of Information and Communication Technology (SOICT), Hanoi University of Science and Technology (HUST); School of Science, Engineering and Technology, Penn State Harrisburg, The Pennsylvania State University（河内科技大学信息与通信技术学院; 宾夕法尼亚州立大学哈里斯堡分校科学、工程与技术学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对联邦学习中非独立同分布数据导致的收敛慢问题，提出FedImp算法，通过量化设备数据信息内容计算聚合权重，在EMNIST和CIFAR-10数据集上显著减少通信轮次并提升性能。
AI 中文摘要
联邦学习（FL）是一种协作范式，允许多个设备在训练全局模型的同时保护本地数据隐私。FL面临的主要挑战是设备间数据的非独立同分布（non-IID）特性，这会降低训练效率、减缓收敛速度。为解决该问题，我们提出联邦杂质加权（FedImp）算法，该算法基于各设备本地数据的信息内容量化其贡献，将这些贡献归一化后计算全局模型更新的不同聚合权重。在EMNIST和CIFAR-10数据集上的大量实验表明，与FedAvg、FedProx、FedAdp相比，FedImp在EMNIST上最多减少64.4%、27.8%、66.7%的通信轮次，在CIFAR-10上最多减少44.2%、44%、25.6%的通信轮次；在高度不平衡的数据分布下，FedImp的性能优于所有基准方法，且达到最高准确率。总体而言，FedImp为非IID场景下提升FL效率提供了有效解决方案。
英文摘要
Federated Learning (FL) is a collaborative paradigm that enables multiple devices to train a global model while preserving local data privacy. A major challenge in FL is the non-Independent and Identically Distributed (non-IID) nature of data across devices, which hinders training efficiency and slows convergence. To tackle this, we propose Federated Impurity Weighting (FedImp), a novel algorithm that quantifies each device contribution based on the informational content of its local data. These contributions are normalized to compute distinct aggregation weights for the global model update. Extensive experiments on EMNIST and CIFAR-10 datasets show that FedImp significantly improves convergence speed, reducing communication rounds by up to 64.4%, 27.8%, and 66.7% on EMNIST, and 44.2%, 44%, and 25.6% on CIFAR-10 compared to FedAvg, FedProx, and FedAdp, respectively. Under highly imbalanced data distributions, FedImp outperforms all baselines and achieves the highest accuracy. Overall, FedImp offers an effective solution to enhance FL efficiency in non-IID settings.
CommentsAccepted author manuscript (AAM) to appear in IEEE Transactions on Artificial Intelligence
Journal refIEEE Transactions on Artificial Intelligence, vol. 7, no. 3, pp. 1652-1665, March 2026
