UNIVERSITY OF HONG KONG
FACULTY OF ARCHITECTURE
Seminars

Yuebing LIANG

Yuebing LIANG
Date:
28-Mar-2022
2:00 pm - 3:00 pm
Venue:
To be held virtually via Zoom
Title:
Generalizable deep neural networks for human mobility prediction: knowledge sharing across cities and modes
Speaker(s):
Miss Yuebing LIANG
PhD Student
Department of Urban Planning and Design
The University of Hong Kong

The Centre of Urban Studies and Urban Planning is pleased to invite you to attend the RPG Research Seminars on 28 March 2022, (Monday) 14:00 to 15:00 via Zoom:

Please login 15 minutes before the session starts.

*Pre-registration required, zoom link will be provided upon successful registration. [Deadline: 25 March 2022 (Friday) 12:00 p.m. (HKT)]
Registration: https://bit.ly/3MY6nAH


Abstract
Human mobility prediction based on urban characteristics plays an important role in city planning. Early studies usually use mathematical models to depict human mobility patterns, which are over-simplified and suffer from low accuracy. Most recently, several attempts have demonstrated the superior performance of deep neural networks for human mobility prediction. Unfortunately, their success is mostly conditioned on large-scale training data in the same scenario, which are often inaccessible in real-world planning stages. The generalizability of models to different scenarios are extremely important for human mobility prediction, not only to deal with the data scarcity problem in the planning of new cities or transport modes, but also to ensure that the models indeed capture true and universal knowledge. For this purpose, this research tries to develop generalizable deep neural networks for human mobility prediction by sharing knowledge across cities and modes. Based on the generalizable model, we can further interpret the relationship between human mobility patterns and urban characteristics, which can provide general strategic guidance for urban planners and policy makers.


About the Speaker
I am Yuebing Liang, a second-year PhD student in urban planning and design at the University of Hong Kong (HKU). Before joining HKU, I got a Master and Bachelor of Architecture degree from Tsinghua University. My research interests lie in the intersection of data science, human mobility and urban science. In particular, my previous research contributes to data-driven human mobility modeling using advanced machine learning techniques.


~~ ALL INTERESTED ARE WELCOME ~~
Enquiries: 3917 2721
CENTRE OF URBAN STUDIES AND URBAN PLANNING
THE UNIVERSITY OF HONG KONG

SEARCH