EVENT DETAILS
Jianwei Huang Huais a Presidential Chair Professor and Associate Vice President (Institutional Development) of the Chinese University of Hong Kong, Shenzhen. He received his Ph.D. from Northwestern University in 2005 and worked as a Postdoc Research Associate at Princeton University from 2005 to 2007. His research interests are network optimization and economics, with applications in communication networks, energy networks, data markets, and crowd intelligence. In the big data era, effective data trading is crucial for enhancing decision-making by machine learning algorithms. Traditional methods, designed for static databases, fail to address the need for continuous query-based trading over streaming data, often resulting in revenue losses. We introduce CQTrade, the first framework for continuous query-based data trading, which optimizes computation sharing across time windows and queries. This innovation maximizes seller revenue by integrating seamlessly with existing trading mechanisms. Our approach includes a theoretical analysis of computation-sharing techniques and an optimization problem solved by a branch-and-price algorithm. CQTrade improves trading success rates by 12.8% and boosts revenue by 28.7% over traditional methods, marking a significant advancement in data trading. This is a joint work with Jin Cheng, Ningning Ding, and John Lui, published in Sigmetrics 2024. At the end of the talk, I will also briefly introduce CUHK Shenzhen, a leading international research university sharing the same brand of CUHK.
TIME Wednesday September 18, 2024 at 11:00 AM - 12:00 PM
LOCATION L440, Technological Institute map it
ADD TO CALENDAR&group= echo $value['group_name']; ?>&location= echo htmlentities($value['location']); ?>&pipurl= echo $value['ppurl']; ?>" class="button_outlook_export">
CONTACT Catherine Healey catherine.healey@northwestern.edu
CALENDAR Department of Electrical and Computer Engineering (ECE)