My research areas are dynamical systems and ergodic theory, theory of machine learning and Bayesian statistics. Broadly, I like mathematical analyses and computational methods that can demystify complex dynamics and be useful for engineering them. I am particularly motivated by fundamental scientific questions about nonlinear, possibly chaotic, dynamical systems arising in climate science, biomedical and machine learning applications. I enjoy thinking about problems in these application areas where some/all of i) domain knowledge-based models, ii) experimental/numerical timeseries data, and iii) other traditional theory are available and can be fruitfully complemented with a dynamical systems approach.

If you are a student looking for research projects, please check the high-level summaries page. Please reach out to me by email if you are interested in learning more!

Background

Prior to joining Georgia Tech, I was a postdoc at MIT in the Institute for Data, Systems and Society. I worked with Stefanie Jegelka and Youssef Marzouk in the areas of machine learning theory and Bayesian statistics respectively. I received my PhD from MIT, where I worked with my advisor, Qiqi Wang, in the field of computational dynamical systems. My Master’s work, under the supervision of Nicolas Hadjiconstantinou, also at MIT, was on nanoscale fluids. Prior to that, I received a Bachelor’s degree in Mechanical Engineering from Indian Institute of Technology, Roorkee. Through undergraduate internships, I studied numerical PDEs and scientific computing from Mary Catherine Kropinski, Praveen Chandrashekar and Karthik Duraisamy. As a graduate student, I did internships in Scott Murman’s group at NASA Ames and with Sri Hari Krishna Narayanan and Paul Hovland at Argonne National Lab. I learned a lot from all the abovementioned wonderful advisors and from my other collaborators, some of whom are co-authors on our publications.