Deep Spatiotemporal Models for Sequence Prediction
Many real-world machine learning applications require learning the association between input and output sequences. Some of these applications involve spatiotemporal sequences in both the input and output. Examples can be found in weather forecasting, visual object tracking, crowd analysis, and many other video processing applications. In this talk, some recent deep spatiotemporal models we have developed in recent years will be presented. While different potential applications will be highlighted, focus will be put on a collaborative project with the Hong Kong Observatory to apply our models to a challenging precipitation nowcasting application.
About Prof. Dit-Yan YEUNG
Dit-Yan Yeung received his BEng degree in electrical engineering and MPhil degree in computer science from the University of Hong Kong and PhD degree in computer science from the University of Southern California. He started his academic career as an assistant professor at the Illinois Institute of Technology in Chicago. He later joined the Hong Kong University of Science and Technology where he is now a professor in the Department of Computer Science and Engineering, with joint appointment in the Department of Electronics and Computer Engineering. He has been doing research in machine learning since his doctoral study, focusing on computational and statistical approaches to machine learning and artificial intelligence.