论文标题
联合积极学习（F-AL）：联合学习的有效注释策略
Federated Active Learning (F-AL): an Efficient Annotation Strategy for Federated Learning
论文作者
论文摘要
从沟通效率，隐私和公平性方面，对联邦学习（FL）进行了深入研究。但是，有效的注释是现实世界中FL应用中的一个痛点，但研究较少。在这个项目中，我们建议将主动学习（AL）和采样策略应用于FL框架，以减少注释工作量。我们希望AL和FL可以互补地提高彼此的表现。在我们提出的联合积极学习（F-AL）方法中，客户协作实施了AL，以获取以分布式优化方式获得FL的实例。我们使用常规的随机抽样策略，客户级单独的AL（S-AL）和提议的F-AL比较了全局FL模型的测试精度。我们从经验上证明，F-AL在图像分类任务中的表现优于基线方法。
Federated learning (FL) has been intensively investigated in terms of communication efficiency, privacy, and fairness. However, efficient annotation, which is a pain point in real-world FL applications, is less studied. In this project, we propose to apply active learning (AL) and sampling strategy into the FL framework to reduce the annotation workload. We expect that the AL and FL can improve the performance of each other complementarily. In our proposed federated active learning (F-AL) method, the clients collaboratively implement the AL to obtain the instances which are considered as informative to FL in a distributed optimization manner. We compare the test accuracies of the global FL models using the conventional random sampling strategy, client-level separate AL (S-AL), and the proposed F-AL. We empirically demonstrate that the F-AL outperforms baseline methods in image classification tasks.
