Short Bio

I am a Computer Science PhD student at Harvard University, and an affiliate at the Harvard/MIT Center for Brains, Minds and Machines. I am advised by Hanspeter Pfister and Xavier Boix, and collaborate closely with Gabriel reiman and Tomaso Poggio. My research focusses on understanding and improving the out-of-distribution generalization in computer vision. I am fortunate to be supported by the Winston Chen Family Graduate Fellowship for funding my research.

Research Goal

Generalizing beyond training data (i.e. out-of-distribution) is the fundamental goal of machine learning. While standardized datasests have lead to unprecedented success in computer vision, understanding how our models generalize beyond them is essential for building robust, trustable models which can be deployed in the society with confidence. My work approaches this problem of understanding and improving generalization by building controlled 3D worlds using Computer Graphics to debug computer vision models. For an updated list of my papers, please check my google scholar page or my CV.

Beyond Research

  • I am an avid guitarist. You can find my covers and original music here - Bandcamp and Soundcloud.
  • I made an android app targeted to hindi users which got over 50,000 downloads!
  • I love to teach and used the pandemic to teach people diverse backgrounds (including lawyers, financial consultants) how to code in Python. Content coming soon!