ECE seminar series: Real-time knowledge discovery and decision making with Big Data
Wednesday, September 10, 2014 – ( 3:00 pm to 4:00 pm )
Location: Babbio Center, Room 220
Real-time knowledge discovery and decision making with Big Data
BY Jie Xu
University of California, Los Angeles
As the world becomes ever more connected and instrumented, decision-makers have ever more rapid access to ever changing and growing streams of data – but this makes the decision-maker’s problems ever more complex as well, because it is impossible to learn everything in the time frame in which decisions must be made. What the decision-maker must do, therefore, is to discover in real time what is relevant in the enormous stream of data and use the relevant information to make good decisions. This talk presents a systematic framework and associated algorithms that enable a decision-maker to do this, and shows how to use them in real-time traffic prediction as an application scenario. With the vast availability of traffic sensors from which traffic information can be derived in real-time, a lot of research effort has been devoted to developing traffic prediction techniques, which in turn improve route navigation, traffic regulation, urban area planning and etc. One key challenge in traffic prediction is how much to rely on prediction models that are constructed using historical data in real-time traffic situations. Our decision framework learns from the current traffic situation in real-time and predicts the future traffic by matching the current situation to the most effective prediction model. When the traffic prediction involves multiple distributed learners but only the feedback about the overall effect of their decisions is available, we also propose fast learning algorithms by exploiting the informativeness of the global feedback. The algorithms we propose yield strong performance guarantees for both the long run and the short run. The applications are numerous besides traffic prediction, including patient monitoring, surveillance, social networks etc.
Jie Xu is a final year PhD student at University of California at Los Angeles. He is advised by Prof. Mihaela van der Schaar in the Department of Electrical Engineering. Prior to attending UCLA, he received his BS and MS degrees in Electrical Engineering from Tsinghua University in China and graduated with honor. His research spans the area of machine learning, data mining and game theory, with an emphasis on learning and incentive design in networks, and has resulted in more than 20 papers published in top journals and conferences in Electrical Engineering and Economics. Jie was the recipient of the best paper award of APCC'2009 and several distinguished fellowships.