Urban spatiotemporal transfer learning: a survey on problems, applications and related techniques

PhD Qualifying Examination


Title: "Urban spatiotemporal transfer learning: a survey on problems, 
applications and related techniques"

by

Mr. Xu GENG


Abstract:

Urban computing has become an active research field in machine learning, 
with various applications to facilitate citizen's life. However, 
traditional machine learning approaches require a large amount of 
high-quality data, which is a big challenge to data acquisition and 
warehousing infrastructure. Also, cities as dynamic systems, always cause 
model expiration problem. Transfer learning is a prominent solution to the 
above problems. In this survey, we investigated transfer learning 
techniques on urban spatiotemporal data, which is the most commonly used 
data type in urban computing. This article surveys state-of-the-art 
transfer learning techniques for cross-city knowledge transfer, temporal 
transfer and other important transfer learning applications that leverage 
spatiotemporal information. In this survey, we define the research 
problems, categorize the existing methodologies and summarize the key 
technical points. This survey further proposes future research 
opportunities for urban computing and urban transfer learning in diverse 
applications. This survey will enable researchers to gain a better 
understanding of the state of urban computing, urban transfer learning and 
spatiotemporal transfer learning and identify the directions for future 
research.


Date:                   Thursday, 13 May 2021

Time:                   4:00pm - 6:00pm

Zoom meeting:           https://hkust.zoom.us/j/6114150123

Committee Members:      Prof. Qiang YANG (Supervisor)
                        Prof. Lei Chen (Supervisor)
                        Prof. Kai Chen (Chairperson)
                        Prof. Xiaojuan Ma