Federated Active Learning (F-AL): an Efficient Annotation Strategy for Federated Learning

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...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2022-02
Main Authors: Jin-Hyun, Ahn, Kim, Kyungsang, Koh, Jeongwan, Li, Quanzheng
Format: Article
Language:English
Subjects:
Active learning
Annotations
Federated learning
Image classification
Optimization
Random sampling
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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.
ISSN:2331-8422
DOI:10.48550/arxiv.2202.00195