Building A TrustworthyCyber World for A Sustainable Physical World

A C2SR Colloquia Series | Distinguished Webinar Series.
The Distinguished Speaker Webinar Series aims to advance state-of-the-art concepts and methods in artificial intelligence and cyber security. It is jointly hosted by the Centers for Cyber Security and AI Research and the School of Electrical Engineering and Computer Science (SEECS) at the University of North Dakota College of Engineering & Mines.
Speaker Biography:
Dr. Guang Wang is an Assistant Professor of the Department of Computer Science at Florida State University. He was a Connection Science fellow and Postdoctoral Associate at the Massachusetts Institute of Technology and obtained his Ph.D. degree in Computer Science from Rutgers University. Guang is interested in finding meaningful patterns from large-scale data and then designing socially informed decision-making algorithms to address real-world societal challenges. Currently, his research is mainly focusing on Cyber-Physical Systems, Spatiotemporal Data Mining, Human-Centered Computing, and Trustworthy Machine Learning (Fairness, Privacy, Transparency), especially for applications on Human Mobility, connected community, Healthcare, Economy, Climate Change, and Sustainability. He has been publishing extensively in top-tier conferences and journals, including 65 papers in Nature Cities, KDD, MobiCom, UbiComp, VLDB, RTSS, ICDE, WWW, etc. He has been honored with the Outstanding Paper Award in the IEEE RTSS 2021.
About the Webinar:
In this talk, I will introduce my research on building a trustworthy cyber world for a sustainable physical world, which technically integrates Generative AI, Spatio-temporal Data Mining, and Real-Time Decision-Making to address real-world scientific and societal challenges in diverse physical systems (e.g., mobility, delivery, and emergency management). In particular, I will first use community road map generation as a concrete scenario, and show how we combine Mobile Sensing and Generative AI to cost-efficiently and automatically generate community road networks on a large scale. I will then show how we design a general uncertainty calibration framework to improve the prediction reliability of any spatiotemporal prediction model. Finally, I will briefly discuss some of our ongoing projects, e.g., privacy-preserving mobility data generation and sharing and climate change mitigation via transportation electrification.