My research broadly focuses on robust and efficient statistical inference with high dimensional and/or incomplete observational data that arise frequently in the modern "big data" era in a variety of scientific disciplines. Most of my work is inspired by real world scientific questions and statistical challenges arising from the analysis of such complex data. Through my work, I try to answer such questions through principled formulation of the underlying statistical problem, and developing methods with provable guarantees and scalable implementations, so that they actually serve a meaningful purpose for practitioners faced with these questions.
Some of my main research interests include: Semi-supervised learning; Causal inference and missing data; Transfer learning; High dimensional semi-parametric inference; Bayesian semi-parametric inference; Applications in large-scale biomedical studies.