IOTSC TALK SERIES: Navigating the Shadows: Unraveling Variable Subset Forecasting

Dear Colleagues and Students,
The State Key Laboratory of Internet of Things for Smart City would like to invite you to join our “IOTSC Talk Series” on 22/03/2024 (Friday). We are pleased to invite Prof. Pengyang WANG from IOTSC and FST as the speaker.

Navigating the Shadows: Unraveling Variable Subset Forecasting
Speaker: Prof. Pengyang WANG, State Key Laboratory of Internet of Things for Smart City and Faculty of Science and Technology, University of Macau
Date: 22/03/2024 (Friday)
Time: 15:00 – 16:00
Language: English
Venue: N21-5007 (Exhibition Hall)

Abstract:
Variable Subset Forecasting (VSF) is a unique and challenging scenario in time series forecasting, where some of the variables presented during the training phase are entirely missing when making inferences. Unlike typical cases of missing data, where only a few ad hoc data points are absent, VSF involves the complete absence of certain variables during the inference phase, forcing the model to rely on a limited subset of the data it was originally trained on. The occurrence of VSF is often observed in contexts such as the Internet of Things (IoT), where sensor networks are prevalent. In this talk, I will introduce the definitions and challenges of VSF, recent works of my group, and future directions.

Speaker’s Bio:
Dr. Pengyang WANG’s research interests are in data mining, machine learning and big data analytics, especially in representation learning for Spatial-Temporal Graph-Structured data, with applications to smart cities, transportation, user/system behavior modeling, critical infrastructure defense, and complex network analysis.

For enquiries: Tel: 8822 9141
Email: shirleyfong@um.edu.mo

Best Regards,
State Key Laboratory of Internet of Things for Smart City