Bio
Ye Wei is currently pursuing his PhD in the Department of Computer Science at Loughborough University, under the supervision of Dr Hui Fang and Dr Haitao He. His research focuses on the complex field of time-series and spatial-temporal forecasting, with a particular emphasis on its application in large-scale urban contexts. In 2022, Wei participated in the Traffic4cast competition, where he managed to place 7th, and subsequently shared his work at the NeurIPS 2022 Traffic4cast workshop.
Wei's academic journey began at Shanghai Jiao Tong University, where he earned his Bachelor’s degree in Electrical and Computer Engineering. Before embarking on his PhD journey, he gained invaluable experience by working as a research assistant in multiple laboratories, including the Advanced Network Laboratory and the Laboratory for Emerging Memory and Low Power Computing at Shanghai Jiao Tong University.
Research interests
Ye Wei's Turing-related research revolves around the enhancement of traffic forecasting, a critical component of Intelligent Traffic Systems (ITS). His objective is to overcome the limitations inherent in traditional knowledge-based and data-driven forecasting methods. These methods often compromise predictive accuracy due to unrealistic assumptions and oversimplifications, and also fail to account for spatio-temporal correlations.
Wei's research harnesses the potential of Graph Neural Networks (GNNs), an advanced deep learning methodology adept at capturing spatial dependencies through non-Euclidean graph structures. Despite GNNs' efficacy in highway systems, their application in urban networks presents difficulties due to sparse sensor data and complex network topologies.
In addition to this, Wei has a keen interest in multivariate time-series forecasting, spatial-temporal forecasting, and mobility dynamics.