Kevin S. XuGraduate Research Assistant |
![]() |
Research | Education | Selected publications | Honors and awards | Recent presentations and posters | Teaching | Curriculum vitae
I am a PhD student in the Electrical Engineering: Systems graduate program in the Electrical Engineering and Computer Science (EECS) department at the University of Michigan. I am currently a graduate research assistant under the supervision of Professor Alfred O. Hero III. I am supported in part by an award from the Natural Sciences and Engineering Research Council of Canada.
My current research interests are in statistical machine learning and network science.
In statistical machine learning, I am particularly interested in methods for learning from time-varying data; that is, identifying patterns in data that are collected over time where the underlying processes generating the data may be non-stationary. I am currently working on problems including tracking and prediction, clustering, and visualization of high-dimensional time-varying data sets.
My main interests in network science also center around the theme of time evolution. Specifically, I am interested in methods to reveal the structure and dynamics of networks as they evolve with time and to detect changes in network structure. I am also interested in statistical modeling and inference on time-evolving networks. My current work is focused on social and information networks, although most of the methodology is also applicable to other types of networks as well.
Key words: machine learning, statistical signal processing, data mining, non-stationary data, tracking and prediction, clustering, visualization, network science, community detection, statistical modeling and inference, social and information networks
Supporting material such as code, data, and presentation slides are available for most publications at the Hero group reproducible research website.