Research

My research focuses on developing high-dimensional and semi-parametric Bayesian methods motivated by improving practical performance in real world applications and studying theoretical properties of such proce- dures. My current and previous work focuses on procedures for high-dimensional Bayesian model selection and developing continuous shrinkage priors on functions in nonparametric settings. I have also worked on a scalable stochastic search algorithm to explore the model space of high-dimensional linear models. In general, I am broadly interested in obtaining theoretically-grounded solutions to applied biomedical and machine learning problems. One of my fundamental research goals is to develop general purpose tools for inferences, predictions, and hypothesis testing in high-dimensional settings, while also developing scalable computing algorithms for implementing such tools. Recently, I have started a new research project for scalable uncertainty quantification via generative processes. The main idea of this is to circumvent a computational bottleneck in traditional frameworks, like posterior computation via MCMC, by constructing a generator of parameter samples from the corresponding posterior distribution.