|Position: Assistant Professor
Office: TBL 309B
E-mail: [email protected]
- Ph.D. Johns Hopkins University 2021
- M.S. Johns Hopkins University 2017
- B.S. Johns Hopkins University 2015
- Causal inference
- Graphical models
- Missing data
Rohit did all of his schooling at Johns Hopkins University. His graduate work spanned topics in causal inference, machine learning, and computational genomics. For his dissertation, Rohit focused on developing practical machine learning methods to separate causation from correlation despite complications such as missing data, confounding variables, and “spillover” of causal effects due to interactions in a social network. At Williams, Rohit teaches courses on topics in causal reasoning, machine learning, and introductory computer science.
Rohit’s research is situated at the intersection of machine learning and statistics. He primarily works on developing principled methods for inferring causality from unstructured and messy data. He is also interested in the application and analysis of these methods in various contexts, such as computational genomics, cancer immunotherapy, and algorithmic fairness. As part of these efforts, Rohit maintains an open-source Python package for causal inference called Ananke.