Viraj Mehta

Ph.D. student at the Robotics Institute at Carnegie Mellon University advised by Jeff Schneider.

prof_pic.jpg
virajm at cs.cmu.edu

Newell-Simon Hall 3128

5000 Forbes Ave

Pittsburgh, PA 15213

As a researcher, I am broadly interested in reinforcement learning, generative models, and dynamical systems. In particular I work on solving control problems in science with machine learning in regimes where the data-generating process is expensive. Much of my work is motivated by the problem of plasma control for nuclear fusion; the various difficulties we face there frequently inspire more general machine learning problems we can solve as computer scientists. My fusion research is conducted in collaboration with the Princeton Plasma Physics Laboratory.

Prior to my time at CMU, I spent time at KKR helping jumpstart their efforts with alternative data and analytics and at Hum Capital (formerly Capital Technologies) doing initial work on automating parts of capital allocation in private finance.

I completed a B.S. in Mathematics and an M.S. in Computer Science at Stanford University, where I conducted research on 3D vision and robot learning advised by Silvio Savarese.

I’m from Austin, Texas and spend my time bouncing between there, Pittsburgh, and sometimes NYC, the Bay Area, and Tahoe. Outside of work, I spend my time flying airplanes, lifting weights, reading, and skiing.

selected publications

  1. Neural Dynamical Systems: Balancing Structure and Flexibility in Physical Prediction
    Mehta, Viraj, Char, Ian, Neiswanger, Willie, Chung, Youngseog, Nelson, Andrew Oakleigh, Boyer, Mark D., Kolemen, Egemen, and Schneider, Jeff
    In IEEE Conference on Decision and Control 2021
    HTML
  2. Representational aspects of depth and conditioning in normalizing flows
    Koehler, Frederic, Mehta, Viraj, and Risteski, Andrej
    In International Conference on Machine Learning 2021
    HTML
  3. An Experimental Design Perspective on Model-Based Reinforcement Learning
    Mehta, Viraj, Paria, Biswajit, Schneider, Jeff, Ermon, Stefano, and Neiswanger, Willie
    In International Conference on Learning Representations 2022
    HTML
  4. Near-optimal Policy Identification in Active Reinforcement Learning
    Li*, Xiang, Mehta*, Viraj, Kirschner*, Johannes, Char, Ian, Neiswanger, Willie, Schneider, Jeff, Krause, Andreas, and Bogunovic, Ilija
    In International Conference on Learning Representations (oral, top 5% of accepted papers) 2023
    HTML
  5. Controlling Plasma Profiles in a Learned Model via Reinforcement Learning
    Mehta, Viraj, Abbate, Joseph, Conlin, William, Kolemen, Egemen, and Schneider, Jeff
    In American Physical Society Division of Plasma Physics Annual Meeting Nov 2021
    HTML
  6. Exploration via Planning for Information about the Optimal Trajectory
    Mehta, Viraj, Char, Ian, Abbate, Joseph, Conlin, Rory, Boyer, Mark D, Ermon, Stefano, Schneider, Jeff, and Neiswanger, Willie
    In Advances in Neural Information Processing Systems 2022
    HTML