Multi-source Transfer Learning with Multi-view Adaboost

Transfer learning, which is one of the most important research directions in machine learning, has been studied in various fields in recent years. In this paper, we combine the theories of multi-source and multi-view learning into transfer learning and propose a new algorithm named Multi-source Transfer Learning with Multi-view Adaboost (MsTL-MvAdaboost). Different from many previous works on transfer learning, in this algorithm, we not only use the labeled data from several source tasks to help learn one target task, but also consider how to transfer them in different views synchronously. We regard all the source and target tasks as a collection of several constituent views and each of these tasks can be learned from different views. Experimental results also validate the effectiveness of our proposed approach.