Research

I am interested in developing principled and practical algorithms for real-world problems. In particular, I am interested in accurately modeling user interactions/requirements, and developing new frameworks for "well-behaved" data and algorithms.

My research has both theoretical and applied components. On the theoretical side, I have worked on "clusterability" assumptions, clustering with limited distance information, clustering with user feedback, and regularization constraints for supervised learning. On the applied side, I have worked on applying theoretical approaches to problems in Bioinformatics, natural language processing, and applications at Google.

My Ph.D. thesis presents algorithms for clustering with limited distance information, and tools for locally exploring large networks.

Ph.D. Thesis: Clustering and Network Analysis with Biological Applications.

Advisors: Shang-Hua Teng, Yu Xia.