About me

I am a fourth year PhD candidate in Computer Science at UMass Amherst, advised by Professor Andrew Lan, with a focus on Machine learning. My PhD research is on improving data selection efficiency with differentiable optimization layers (and avoid heuristic informativeness measures). Recently, I won the NeurIPS 2020 education challenge for the task of improving question selection efficiency in personalized tests (e.g., GRE, GMAT). I also work on robust models for learning with limited, weakly supervised, and noisy labels. During my PhD, I was able to squeeze in two research internships at Adobe Research, advised Dr. Saayan Mitra, Dr. Somdeb Sarkhel, and Dr. Vishy Swaminathan, where I was fortunate to work on optimal bidding strategy in real-time bidding systems. Before my PhD, I have been a Software Engineer (Applied Scientist) at the Microsoft BingAds division where I had the opportunity to work on improving click performance and ad selection algorithms, working closely with Rahul Agrawal. Even before that, I had amazing years in the Indian Institute of Science, Bangalore (IISc), advised by Professor P.S. Sastry, where I had started working on robust learning under noisy labels. I have been fortunate to work with some amazing researchers and publish in many exciting machine learning/data mining conferences and workshops such as AAAI, CVPR, ICML, IJCAI, KDD, SIAM SDM. I also received the Best Student Paper Award from the IEEE Big Data (2020) conference.

Research Interest

  • Meta Learning, Differentiable Optimization Layers, Bi-level Optimization
  • Sequential Models, Reinforcement Learning, Inverse Reinforcement Learning
  • Computational Advertising, Computer Vision, NLP, AI for Social Goods


  • April 2021: paper on computerized adaptive testing under bilevel framework has been accepted at IJCAI 2021 (13.9% Acceptance rate!).
  • April 2021: paper on learning with noisy labels with contrastive initialization to appear at LLID Workhop at CVPR 2021.
  • Dec 2020: honored to get the Best Student Paper Award at IEEE Big Data 2020.
  • Nov 2020: won the personalized question selection task in NeurIPS 2020 Education Challenge. Implementation is publicly available.
  • Nov 2020: paper on robust sample reweighting strategy without gold samples to appear at WACV 2021. Paper/codes will be released soon.
  • Nov 2020: paper on optimal career trajectory modeling to appear at IEEE Big Data 2020. Paper/codes will be released soon.
  • Sep 2020: I will be serving as Program committee member for AAAI 2021 and WACV 2021 .
  • Aug 2020: selected for KDD 2020 Student Award!
  • May 2020: paper on knowledge tracing to appear at KDD 2020. Paper/codes/data will be released soon.
  • Mar 2020: selected for SIAM SDM Student Travel Award.
  • Dec 2019: paper on optimal bidding strategy to appear at SIAM SDM 2020.
  • May 2019: paper on point processes to appear at ICML Time series workshop 2019.
  • May 2019: I will be joining Adobe Research, San Jose for internship (again).
  • Sep 2018: Had a great summer at Adobe Research, San Jose.
  • Aug 2017: Last day at Microsoft!
  • Jul 2017: I will be joining UMass Amherst for PhD in Computer Science.
  • Feb 2017: paper on robust loss functions for deep networks to appear at AAAI 2017.