A Survey on Recent Advances of Transfer Learning

PhD Qualifying Examination


Title: "A Survey on Recent Advances of Transfer Learning"

by

Mr. Ben TAN


Abstract:

Traditional machine learning algorithms always suffer great performance 
loss when the training and test data follow different distribution. In 
practice, this problem commonly exists. Transfer Learning, which aims at 
alleviating the performance degradation on this distribution shift 
occasion, has emerged as one of the most important learning paradigm in 
machine learning and received continuous popularity in research and 
industrial communities. In this paper, we survey some recent transfer 
learning algorithms and their applications. We discuss their merits and 
drawbacks, and identify some potential research problems.


Date:			Tuesday, 10 November 2015

Time:                  	4:00pm - 6:00pm

Venue:                  Room 5560
                         Lifts 27/28

Committee Members:	Prof. Qiang Yang (Supervisor)
 			Prof. Fangzhen Lin (Chairperson)
 			Dr. Wilfred Ng
 			Prof. Dit-Yan Yeung


**** ALL are Welcome ****