{"title":"Stabilizing and improving federated learning with highly non-iid data and client dropout","authors":"Jian Xu, Meilin Yang, Wenbo Ding, Shao-Lun Huang","doi":"10.1007/s10489-024-05956-3","DOIUrl":null,"url":null,"abstract":"<div><p>The label distribution skew has been shown to be a significant obstacle that limits the model performance in federated learning (FL). This challenge could be more serious when the participating clients are in unstable network circumstances and drop out frequently. Previous works have demonstrated that the classifier head is particularly sensitive to the label skew. Therefore, maintaining a balanced classifier head is of significant importance for building a good and unbiased global model. To this end, we propose a simple yet effective framework by introducing a calibrated softmax function with smoothed prior for computing the cross-entropy loss, and a prototype-based feature augmentation scheme to re-balance the local training, which provide a new perspective on tackling the label distribution skew in FL and are lightweight for edge devices and can facilitate the global model aggregation. With extensive experiments on two benchmark classification tasks of Fashion-MNIST and CIFAR-10, our numerical results demonstrate that our proposed method can consistently outperform the baselines, 2 8% of accuracy over FedAvg in the presence of severe label skew and client dropout.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 3","pages":""},"PeriodicalIF":3.4000,"publicationDate":"2024-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Applied Intelligence","FirstCategoryId":"94","ListUrlMain":"https://link.springer.com/article/10.1007/s10489-024-05956-3","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0
Abstract
The label distribution skew has been shown to be a significant obstacle that limits the model performance in federated learning (FL). This challenge could be more serious when the participating clients are in unstable network circumstances and drop out frequently. Previous works have demonstrated that the classifier head is particularly sensitive to the label skew. Therefore, maintaining a balanced classifier head is of significant importance for building a good and unbiased global model. To this end, we propose a simple yet effective framework by introducing a calibrated softmax function with smoothed prior for computing the cross-entropy loss, and a prototype-based feature augmentation scheme to re-balance the local training, which provide a new perspective on tackling the label distribution skew in FL and are lightweight for edge devices and can facilitate the global model aggregation. With extensive experiments on two benchmark classification tasks of Fashion-MNIST and CIFAR-10, our numerical results demonstrate that our proposed method can consistently outperform the baselines, 2 8% of accuracy over FedAvg in the presence of severe label skew and client dropout.
期刊介绍:
With a focus on research in artificial intelligence and neural networks, this journal addresses issues involving solutions of real-life manufacturing, defense, management, government and industrial problems which are too complex to be solved through conventional approaches and require the simulation of intelligent thought processes, heuristics, applications of knowledge, and distributed and parallel processing. The integration of these multiple approaches in solving complex problems is of particular importance.
The journal presents new and original research and technological developments, addressing real and complex issues applicable to difficult problems. It provides a medium for exchanging scientific research and technological achievements accomplished by the international community.