Federated Active Learning (F-AL): an Efficient Annotation Strategy for Federated Learning
This addresses annotation efficiency for federated learning applications, but it is incremental as it combines existing techniques.
The paper tackles the problem of inefficient annotation in federated learning by proposing Federated Active Learning (F-AL), which integrates active learning into the federated framework to reduce annotation workload; it empirically shows that F-AL outperforms baseline methods like random sampling and separate AL in image classification tasks.
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.