<thead id="7vlv5"><ruby id="7vlv5"></ruby></thead>
<cite id="7vlv5"></cite><thead id="7vlv5"><dl id="7vlv5"><th id="7vlv5"></th></dl></thead>
<thead id="7vlv5"><ruby id="7vlv5"></ruby></thead>
<ins id="7vlv5"></ins>
<cite id="7vlv5"><dl id="7vlv5"></dl></cite>
<var id="7vlv5"></var>
<menuitem id="7vlv5"></menuitem>
<listing id="7vlv5"></listing><thead id="7vlv5"><dl id="7vlv5"></dl></thead>
<listing id="7vlv5"><del id="7vlv5"></del></listing>

清華大學|信息學院|國家實驗室|English Version

【2018學術報告08】Active Learning for Rare Event Detection in Massive Data Streams



Title: Active Learning for Rare Event Detection in Massive Data Streams

Speaker: Qing Zhao, Professor, Cornell University

Time: 10:00 a.m., 19th, Jul. (Thu.) 

Place: 1-315, FIT Building 

Language: English

Organizer: Research Institute of Information Technology (RIIT), Tsinghua University



Qing Zhao received the Ph.D. degree in electrical engineering from Cornell University, Ithaca, NY, in 2001. She is a Professor and Gordon Lankton Sesquicentennial Faculty Fellow in the School of Electrical and Computer Engineering, Cornell University. Prior to joining Cornell University in 2015, she was a Professor at the University of California, Davis. Her research interests include sequential decision theory and stochastic optimization, machine learning, statistical inference, and algorithmic theory with applications in infrastructure and communication systems and social economic networks. She is a Fellow of IEEE and the recipient of the 2010 IEEE Signal Processing Magazine Best Paper Award and the 2000 Young Author Best Paper Award from the IEEE Signal Processing Society. While on the faculty of UC Davis, she held the title of UC Davis Chancellor’s Fellow.


The problem of detecting rare events of interest in massive data streams and large complex networks is ubiquitous. The rare events may represent opportunities with exceptional returns or anomalies associated with high costs or potential catastrophic consequences. In this talk, we lay out an active inference and learning approach to rare event detection when the total number of hypotheses is large, the observations are noisy, and the prior knowledge on the rare events may be as little as “they are different from the nominal.”