A Comprehensive Survey on Transfer Learning

TLDR

Transfer learning improves target learner performance by transferring knowledge from related source domains, reduces the need for large target datasets, and has become a popular area, yet existing surveys are fragmented and miss recent advances. This survey seeks to comprehensively review and systematize transfer learning research, linking studies and summarizing mechanisms and strategies to clarify the field’s current status and future directions. It reviews more than 40 representative homogeneous transfer learning approaches from data and model perspectives, briefly discusses applications, and evaluates over 20 models on Amazon Reviews, Reuters‑21578, and Office‑31 to illustrate performance differences. The experiments demonstrate that selecting appropriate transfer learning models is essential for different applications, underscoring the importance of model choice in practice.

Abstract

Transfer learning aims at improving the performance of target learners on target domains by transferring the knowledge contained in different but related source domains. In this way, the dependence on a large number of target-domain data can be reduced for constructing target learners. Due to the wide application prospects, transfer learning has become a popular and promising area in machine learning. Although there are already some valuable and impressive surveys on transfer learning, these surveys introduce approaches in a relatively isolated way and lack the recent advances in transfer learning. Due to the rapid expansion of the transfer learning area, it is both necessary and challenging to comprehensively review the relevant studies. This survey attempts to connect and systematize the existing transfer learning research studies, as well as to summarize and interpret the mechanisms and the strategies of transfer learning in a comprehensive way, which may help readers have a better understanding of the current research status and ideas. Unlike previous surveys, this survey article reviews more than 40 representative transfer learning approaches, especially homogeneous transfer learning approaches, from the perspectives of data and model. The applications of transfer learning are also briefly introduced. In order to show the performance of different transfer learning models, over 20 representative transfer learning models are used for experiments. The models are performed on three different data sets, that is, Amazon Reviews, Reuters-21578, and Office-31, and the experimental results demonstrate the importance of selecting appropriate transfer learning models for different applications in practice.