Federated Active Learning

Federated active learning (FAL) aims to efficiently train machine learning models collaboratively across multiple decentralized devices while minimizing the need for expensive and time-consuming data annotation. Current research focuses on developing algorithms that effectively select the most informative data points for labeling, addressing challenges like non-independent and identically distributed (non-IID) data and domain shifts across devices, often employing ensemble methods and uncertainty quantification techniques. This approach is particularly significant for applications like medical image analysis where data is scarce, privacy is paramount, and efficient annotation strategies are crucial for model performance.

Papers

September 6, 2024

Active-Passive Federated Learning for Vertically Partitioned Multi-view Data
Jiyuan Liu, Xinwang Liu, Siqi Wang, Xingchen Hu, Qing Liao, Xinhang Wan, Yi Zhang, Xin Lv, Kunlun He
Multi View Generating User Activity Model Inference Unlabeled Client Federated Active Learning

June 17, 2024

Federated Active Learning Framework for Efficient Annotation Strategy in Skin-lesion Classification
Zhipeng Deng, Yuqiao Yang, Kenji Suzuki
Active Learning Skin Lesion Annotation Strategy Dermoscopic Image Federated Active Learning

December 5, 2023

Think Twice Before Selection: Federated Evidential Active Learning for Medical Image Analysis with Domain Shifts
Jiayi Chen, Benteng Ma, Hengfei Cui, Yong Xia
Active Learning Domain Shift Medical Image Analysis View Selection Evidential Model Federated Active Learning

December 4, 2023

Federated Active Learning for Target Domain Generalisation
Razvan Caramalau, Binod Bhattarai, Danail Stoyanov
Active Learning Federated Learning Domain Generalization Federated Active Learning

March 22, 2023

Re-thinking Federated Active Learning based on Inter-class Diversity
SangMook Kim, Sangmin Bae, Hwanjun Song, Se-Young Yun
Active Learning Local Model Selection Policy Class Diversity Federated Active Learning

November 24, 2022

Knowledge-Aware Federated Active Learning with Non-IID Data
Yu-Tong Cao, Ye Shi, Baosheng Yu, Jingya Wang, Dacheng Tao
Raw Data Decentralized Learning Federated Active Learning

February 1, 2022

Federated Active Learning (F-AL): an Efficient Annotation Strategy for Federated Learning
Jin-Hyun Ahn, Kyungsang Kim, Jeongwan Koh, Quanzheng Li
Active Learning Federated Learning Annotation Strategy Efficient Annotation Sampling Strategy Federated Active Learning