One-shot federated learning (OSFL) has attracted increasing attention for its ability to substantially reduce communication overhead by limiting client-server interaction to a single communication round. Nevertheless, this limited interaction makes learning from heterogeneous client distributions challenging. Diffusion-based OSFL approaches seek to address this challenge by synthesizing data that approximate client distributions. However, existing diffusion-based approaches often impose substantial computational costs on clients and lack effective visual guidance to generate data aligned with client distributions. To address these limitations, we propose FedDoG, a training-free Domain Generation framework that leverages server-side pre-trained diffusion models to synthesize data aligned with client domains. FedDoG extracts compact domain representations using a frozen image encoder, substantially reducing client-side computational and communication costs. These representations are incorporated into the diffusion model through a pre-trained image-conditioning adapter, thereby providing explicit visual guidance for domain generation. To further improve the quality of synthetic data, FedDoG adapts the strength of visual conditioning according to domain concentration and filters generated samples based on class consistency and predictive entropy. Experiments on three benchmarks demonstrate that FedDoG generates higher-quality synthetic images and outperforms existing OSFL baselines in classification accuracy. Meanwhile, FedDoG reduces client upload and computation time by 99.9% and 91.5%, respectively. The code is available at https://github.com/YXTong17/FedDoG.
Collaborative Split Federated Learning with Parallel Training and Aggregation
Department of Informatics, Athens University of Economics and Business(信息学院,雅典经济与商业大学);Munich Research Center, Huawei Technologies Duesseldorf GmbH(慕尼黑研究中心,华为技术杜塞尔多夫有限公司)
FedSSO: A Federated Server-Side Second-Order Optimization Algorithm
Yidu Cloud Technology Inc.(医渡云科技有限公司);Institute for AI Industry Research, Tsinghua University(清华大学智能产业研究院);Army Engineering University of PLA(中国人民解放军陆军工程大学)
Enhancing Robustness of Federated Learning via Server Learning
National Institute of Standards and Technology (NIST)(美国国家标准与技术研究院);University of Maryland, College Park(马里兰大学帕克分校);Tata Consultancy Services Research(塔塔咨询服务公司研究院);University of North Carolina at Charlotte(北卡罗来纳大学夏洛特分校)
An Upload-Efficient Scheme for Transferring Knowledge From a Server-Side Pre-trained Generator to Clients in Heterogeneous Federated Learning
Shanghai Jiao Tong University(上海交通大学);Institute for AI Industry Research (AIR), Tsinghua University(清华大学人工智能产业研究院);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);Queen’s University Belfast(贝尔法斯特女王大学)
Federated Learning via Indirect Server-Client Communications
University of Miami(迈阿密大学);University of Virginia(弗吉尼亚大学)
联邦学习(FL)是一种通信高效、隐私保护的分布式机器学习框架,近年来得到了大量的研究关注。尽管FL算法的形式不同(例如,同步FL、异步FL)和底层优化方法,几乎所有现有的工作都隐含地假设了通信基础设施的存在,该通信基础设施促进了服务器和客户端之间用于模型数据交换的直接通信。然而,这种假设在许多可以受益于分布式学习但缺乏适当的通信基础设施(例如,在偏远地区的智能传感)。在本文中,我们提出了一种新的FL框架,名为FedEx(FL via Model Express Delivery的缩写),它利用移动运输工具(例如,无人机)来建立服务器和客户端之间的间接通信信道。两个算法,称为联邦快递同步和联邦快递异步,开发取决于是否采用同步或同步调度的运输商。即使间接通信引入异构延迟的客户端的全球模型传播和本地模型收集,我们证明了联邦快递的两个版本的收敛。收敛性分析随后揭示了如何将客户端分配给不同的传输器并设计客户端之间的路由。通过在两个公共数据集上的模拟网络中的实验来评估联邦快递的性能。
Efficient Federated Learning Using Dynamic Update and Adaptive Pruning with Momentum on Shared Server Data
Hithink RoyalFlush Information Network Co., Ltd.(同花顺网络信息股份有限公司);School of Computer Science and Technology, Soochow University(苏州大学计算机科学与技术学院);Collaborative Innovation Center of Novel Software Technology and Industrialization(新型软件技术与产业化协同创新中心);Baidu Research(百度研究院);Software Institute, Peking University(北京大学软件学院);Auburn University(奥本大学);North Carolina State University(北卡罗来纳州立大学);School of Computer Science, Fudan University(复旦大学计算机科学学院)
Federated Knowledge Transfer Fine-tuning Large Server Model with Resource-Constrained IoT Clients
School of Intelligent Systems Engineering, Sun Yat-Sen University(中山大学智能工程学院);Shenzhen Fangle Technology Co., Ltd.(深圳方乐科技有限公司);School of Computer Science and Engineering, Nanyang Technological University(南洋理工大学计算机科学与工程学院);School of Computer Science and Technology, Dalian University of Technology(大连理工大学计算机科学与技术学院);School of Electronic Engineering and Computer Science, Queen Mary University of London(伦敦玛丽女王大学电子工程与计算机科学学院)
Perfectly Accurate Membership Inference by a Dishonest Central Server in Federated Learning
Technische Universität Wien(维也纳技术大学);CSC-CONICET(阿根廷国家科学技术研究委员会);Universidad de Buenos Aires(布宜诺斯艾利斯大学);MILA - Quebec AI Institute(魁北克人工智能研究所);CNRS(法国国家科学研究中心);CentraleSupélec(中央高等电力学院)
Server-Enforced Watermarking in U-Shaped Split Federated Learning
Software Engineering Institute, East China Normal University(东华大学软件工程学院);College of Computer Science and Electronic Engineering, Hunan University(湖南大学计算机科学与电子工程学院);College of Control Science and Engineering, China University of Petroleum (East China)(中国石油大学(华东)控制科学与工程学院)
dFLMoE: Decentralized Federated Learning via Mixture of Experts for Medical Data Analysis
School of Software and Microelectronics, Peking University(北京大学软件与微电子学院);PKU-OCTA Laboratory for Blockchain and Privacy Computing(区块链与隐私计算实验室);National Engineering Research Center for Software Engineering, Peking University(软件工程国家工程研究中心);State University of New York at Buffalo(纽约州立大学布法罗分校)
Capture Global Feature Statistics for One-Shot Federated Learning
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络空间安全学院);State Key Laboratory of Cyberspace Security Defense(网络空间安全防御国家重点实验室)