My research interests are machine learning and statistical learning. My advisor is Dr. Milos Hauskrecht.
Active Learning & Group Learning: My thesis work aims to learn accurate classifiers from minimized human supervision. Our approach is called Group Active Learning which learns classifiers from labeled groups in an active online learning manner. A group represents a population of instances; and a group label, which is provided by a human oracle, will summarize the class proportions of the enclosed instances. The essential advantage of such labeling is that through only one group label human can express their belief on classifying a population of instances.
Clinical Machine Learning System Building: This is an engineering-wise project which is funded by National Institutes of Health (NIH). The goal of this project is to build a real-time alerting system which detects outliers in ICU clinical events. The data are from UPMC hospitals. Please see Outlier-based clinical monitoring and alerting
