Object detection with on-board sensors (e.g., lidar, radar, and camera) play a crucial role in autonomous driving (AD), and these sensors complement each other in modalities. While crowdsensing may potentially exploit these sensors (of huge quantity) to derive more comprehensive knowledge, \textit{federated learning} (FL) appears to be the necessary tool to reach this potential: it enables autonomous vehicles (AVs) to train machine learning models without explicitly sharing raw sensory data. However, the multimodal sensors introduce various data heterogeneity across distributed AVs (e.g., label quantity skews and varied modalities), posing critical challenges to effective FL. To this end, we present AutoFed as a heterogeneity-aware FL framework to fully exploit multimodal sensory data on AVs and thus enable robust AD. Specifically, we first propose a novel model leveraging pseudo-labeling to avoid mistakenly treating unlabeled objects as the background. We also propose an autoencoder-based data imputation method to fill missing data modality (of certain AVs) with the available ones. To further reconcile the heterogeneity, we finally present a client selection mechanism exploiting the similarities among client models to improve both training stability and convergence rate. Our experiments on benchmark dataset confirm that AutoFed substantially improves over status quo approaches in both precision and recall, while demonstrating strong robustness to adverse weather conditions.
H2Tune: Federated Foundation Model Fine-Tuning with Hybrid Heterogeneity
School of Artificial Intelligence, Beihang University（北京航空航天大学人工智能学院）；School of Computer Science and Technology, Heilongjiang University（黑龙江大学计算机科学与技术学院）；School of Computer Science and Engineering, Beihang University（北京航空航天大学计算机科学与工程学院）；School of Statistics, Renmin University of China（中国人民大学统计学院）；Zhongguancun Laboratory, China（中关村实验室）；School of Computer Science and Technology, Shandong University（山东大学计算机科学与技术学院）；WeBank Co., Ltd, Shenzhen, China（WeBank有限公司，深圳，中国）；Beijing Academy of Blockchain and Edge Computing（北京区块链与边缘计算研究院）
SCAFFLSA: Taming Heterogeneity in Federated Linear Stochastic Approximation and TD Learning
CMAP, UMR 7641, École polytechnique（法国高等理工学院CMAP研究所）；HSE University（俄罗斯高等经济大学）；Technology Innovation Institute（技术创新研究所）；Steklov Mathematical Institute of Russian Academy of Sciences（俄罗斯科学院斯捷克洛夫数学研究所）
Handling Spatial-Temporal Data Heterogeneity for Federated Continual Learning via Tail Anchor
Southwestern University of Finance and Economics（西南财经大学）；University of Electronic Science and Technology of China（电子科技大学）；WeBank（微众银行）；Southwest Jiaotong University（西南交通大学）；Hong Kong University of Science and Technology（香港科技大学）
An Architecture Built for Federated Learning: Addressing Data Heterogeneity through Adaptive Normalization-Free Feature Recalibration
Department of Computer Science, City St. George’s, University of London（伦敦大学城市圣乔治学院计算机科学系）；BioMedIA, Department of Computing, Imperial College London（伦敦帝国理工学院计算系BioMedIA）
Where to Begin? On the Impact of Pre-Training and Initialization in Federated Learning
Meta AI
联邦学习的一个经常提到的挑战是异质性的存在。\emph{Data heterogeneity}是指来自不同客户端的数据可能遵循非常不同的分布。\emph{System heterogeneity}是指具有不同系统能力的客户端设备。大量的联邦优化方法可以解决这一挑战。在文献中，经验评估通常从随机初始化开始联邦训练。然而，在联邦学习的许多实际应用中，服务器可以访问训练任务的代理数据，这些数据可以用于在开始联合训练之前预训练模型。使用四个标准的联邦学习基准数据集，我们实证研究了从联邦学习中的预训练模型开始的影响。不出所料，从预先训练的模型开始减少了达到目标出错率所需的训练时间，并且能够训练比从随机初始化开始时更准确的模型（高达40%）。令人惊讶的是，我们还发现，从预先训练的初始化开始联邦学习可以减少数据和系统异构性的影响。我们建议未来的工作提出和评估联邦优化方法，以评估从随机和预先训练的初始化开始时的性能。这项研究提出了几个问题，进一步了解联邦优化中的异构性的作用。\footnote{Our code is available at: \url{https://github.com/facebookresearch/where_to_begin}}
Benchmarking Federated Learning for Semantic Datasets: Federated Scene Graph Generation
Graduate School of Artificial Intelligence, Ulsan National Institute of Science and Technology（人工智能研究生院，乌山国立科学技术研究所）；Department of Electrical Engineering, Ulsan National Institute of Science and Technology（电气工程系，乌山国立科学技术研究所）；Electronics and Telecommunications Research Institute (ETRI)（电子电信研究院（ETRI））
Robust Federated Learning on Edge Devices with Domain Heterogeneity
Department of Computer Science and Engineering, Kyung Hee University, 171-04, Republic of Korea（韩国庆熙大学计算机科学与工程系）；Machine learning department, Mohamed Bin Zayed University of Artificial Intelligence, United Arab Emirates（阿联酋穆罕默德·本·扎耶德人工智能大学机器学习系）
Personalized Federated Dictionary Learning for Modeling Heterogeneity in Multi-site fMRI Data
Tulane Center for Biomedical Informatics and Genomics（Tulane生物医学信息学与基因组学中心）；School of Medicine, Tulane University（Tulane大学医学院）；State University of New York at Oswego（纽约州立大学奥尔巴尼分校）
FedNoRo: Towards Noise-Robust Federated Learning by Addressing Class Imbalance and Label Noise Heterogeneity
Huazhong University of Science and Technology（华中科技大学）；Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；Hong Kong University of Science and Technology（香港科技大学）
FedSCAM (Federated Sharpness-Aware Minimization with Clustered Aggregation and Modulation): Scam-resistant SAM for Robust Federated Optimization in Heterogeneous Environments
Department of Computer Science, Lahore University of Management Sciences, Lahore, Pakistan（计算机科学系，拉合尔管理科学大学，拉合尔，巴基斯坦）
FedCoSR: Personalized Federated Learning with Contrastive Shareable Representations for Label Heterogeneity in Non-IID Data
Department of Data Science and AI, Faculty of IT and Monash Energy Institute, Monash University（数据科学与人工智能系，IT学院和莫纳什能源研究所，莫纳什大学）；School of Computer Science and Technology, University of Electronic Science and Technology of China (UESTC), Chengdu, and Shenzhen Institute for Advanced Study of UESTC, Shenzhen, China（计算机科学与技术学院，电子科技大学（UESTC）成都校区，以及UESTC深圳高级研究所，深圳，中国）
FedXDS: Leveraging Model Attribution Methods to counteract Data Heterogeneity in Federated Learning
Fraunhofer Heinrich Hertz Institute（弗劳恩霍夫海因里希·赫兹研究所）；Technical University of Berlin（柏林工业大学）；Berlin Institute for the Foundations of Learning and Data (BIFOLD)（柏林学习与数据基础研究所）
Virtual Homogeneity Learning: Defending against Data Heterogeneity in Federated Learning
Hong Kong Baptist University（香港浸会大学）；The Hong Kong University of Science and Technology（香港科技大学）；The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））
FedMLP: Federated Multi-Label Medical Image Classification under Task Heterogeneity
School of Electronic Information and Communications, Huazhong University of Science and Technology（华中科技大学电子信息与通信学院）；School of Engineering, Hong Kong University of Science and Technology（香港科技大学工程学院）
FedCCA: Client-Centric Adaptation against Data Heterogeneity in Federated Learning on IoT Devices
Department of Computing, The Hong Kong Polytechnic University（香港理工大学计算机系）；Centre for Learning, Teaching, and Technology, The Education University of Hong Kong（香港教育大学学习、教学与技术中心）
Combating Exacerbated Heterogeneity for Robust Models in Federated Learning
Hong Kong Baptist University（香港浸会大学）；Shanghai Jiao Tong University（上海交通大学）；Shanghai AI Laboratory（上海人工智能实验室）；The University of Sydney（悉尼大学）；Tsinghua University（清华大学）
Towards Heterogeneity-Aware and Energy-Efficient Topology Optimization for Decentralized Federated Learning in Edge Environment
School of Engineering, Swinburne University of Technology（工程学院，斯威本科技大学）；School of Computer Science and Technology, Tongji University（计算机科学与技术学院，同济大学）；State Key Lab for Novel Software Technology, Nanjing University（新型软件技术国家重点实验室，南京大学）；School of Computer Science and Artificial Intelligence, Wuhan University of Technology（计算机科学与人工智能学院，武汉科技大学）；School of Cyber Science and Engineering, Wuhan University（网络科学与工程学院，武汉大学）；CSIRO Data61
