Research Vision: Develop trustworthy machine learning methods that are robust to imperfect distributional and computational assumptions.

1. Can causality help us understand and mitigate ML robustness issues?
2. What is the interplay between ML explanations and robustness?
3. How can we perform robust collaborative learning on a dynamic network of edge devices?
4. Can we find robust distribution matching methods to alleviate distribution shifts?

Other Details: I am an assistant professor in Purdue ECE. At CMU, my postdoc advisor was Prof. Pradeep Ravikumar. At UT-Austin, my PhD advisors were Prof. Inderjit Dhillon and Prof. Pradeep Ravikumar. My work has been funded by NSF, ARL and ONR.