A Comprehensive Survey of Federated Transfer Learning: Challenges, Methods and Applications

• DOI: 10.48550/arxiv.2403.01387
• Publication Date: 2024-03-02
• Authors: Wei Guo; Fuzhen Zhuang; Xiao Zhang; Yiqi Tong; Jin Dong
• Keywords: Transfer of learning

Federated learning (FL) is a novel distributed machine paradigm that enables participants to collaboratively train centralized model with privacy preservation by eliminating the requirement of data sharing. In practice, FL often involves multiple and requires third party aggregate global information guide update target participant. Therefore, many methods do not work well due training test each participant may be sampled from same feature space underlying distribution. Meanwhile, differences in their local devices (system heterogeneity), continuous influx online (incremental data), labeled scarcity further influence performance these methods. To solve this problem, federated transfer (FTL), which integrates (TL) into FL, has attracted attention numerous researchers. However, since share knowledge among communication round while allowing accessed other participants, FTL faces unique challenges are present TL. survey, we focus on categorizing reviewing current progress learning, outlining corresponding solutions applications. Furthermore, common setting scenarios, available datasets, significant related research summarized survey.