Federated transfer learning: Concept and application
Tengjun Ni · Applied and Computational Engineering · 2023
As the development of Machine Learning, more and more data are used to train models in order to guide people’s daily life. Faced with privacy and efficiency challenges these days, Federated Transfer Learning (FTL) has grown in popularity, as it has a great ability to protect data privacy while also dealing with the problem of data scarcity. In this passage, the author studies the FTL and its different applications, including sales industry, Industrial Internet of Things Devices, Finance application, Medical Application and Autonomous Driving, analyses the usage in the different industries. Further, the author discusses the privacy and robustness of FTL, which are the core features of FTL. In the end, the author concluded the features of FTL and also gave a prospection of FTL.