Sushmita Bhattacharya

Sushmita

I am a postdoctoral researcher at REACT lab at Harvard Univerisity with Prof. Stephanie Gil. My research focuses on designing multiagent reinforcement learning algorithms to solve real-world problems under computation and communication constraints. As a part of ProjectCETI, I work on model-based RL algorithms to enable autonomous marine life observation.

I completed my Ph.D. in Computer Science from Harvard University, where I worked with Prof. Stephanie Gil on Sequential Decision-Making for Multi-Robot Systems with Real-World Uncertainty using Rollout-based Reinforcement Learning. I recipient of the 2022 Apple Scholars in AI/ML PhD fellowship in Fundamentals of Machine Learning.

  • Resume - October 2024
  • Google Scholar
  • Email
  • Harvard website
  • Linkedin

    • Awards

    • I am a recipient of the 2022 Apple Scholars in AI/ML PhD fellowship in Fundamentals of Machine Learning.
    • I received the Engineering Graduate Fellowship from Ira A. Fulton Schools of Engineering (Spring 2020) for extraordinary academic achievements.

    Publications

    • Reinforcement Learning-Based Framework for Whale Rendezvous via Autonomous Sensing Robots


      Ninad Jadhav*, Sushmita Bhattacharya*, Daniel Vogt, Yaniv Aluma, Pernille Tonessen, Akarsh Prabhakara, Swarun Kumar, Shane Gero, Robert J. Wood, and Stephanie Gil.
      (*equal contribution)
      Science Robotics, 2024
      [ Paper ]

    • Multiagent Reinforcement Learning: Rollout and Policy Iteration for POMDP with Application to Multi-Robot Problems


      Sushmita Bhattacharya, Siva Kailas, Sahil Badyal, Stephanie Gil, Dimitri Bertsekas
      IEEE Transactions on Robotics, 2023
      [ Paper ]

    • Approximate Multiagent Reinforcement Learning for On-Demand Urban Mobility Problem on a Large Map


      Daniel Garces, Sushmita Bhattacharya, Dimitri Bertsekas, Stephanie Gil
      International Conference on Robotics and Automation (ICRA), 2024
      [ Paper ] [ arXiv ]

    • Multiagent Reinforcement Learning for Autonomous Routing and Pickup Problem with Adaptation to Variable Demand


      Daniel Garces, Sushmita Bhattacharya, Stephanie Gil, Dimitri Bertsekas
      International Conference on Robotics and Automation (ICRA), 2023
      [ Paper ] [ arXiv ] [ Video ]

    • Multiagent Rollout and Policy Iteration for POMDP with Application to Multi-Robot Repair Problems


      Sushmita Bhattacharya, Siva Kailas, Sahil Badyal, Stephanie Gil, and Dimitri Bertsekas
      Conference on Robot Learning (CoRL), 2020
      [ Paper ] [ Talk ] [ Video ]

    • Reinforcement Learning for POMDP: Rollout and Policy Iteration with Application to Autonomous Sequential Repair Problems


      Sushmita Bhattacharya, Sahil Badyal, Thomas Wheeler, Stephanie Gil, and Dimitri Bertsekas
      IEEE Robotics and Automation Letters (RA-L), 2020
      [ Paper ] [ arXiv ] [ Video ]

    Media Coverage


    Teaching Assistantship

    • Topics in Reinforcement Learning - ASU Spring 2020 taught by Dr. Dimitri Bertsekas
    • Coordination of Multi-Robot Systems - ASU Fall 2019 taught by Dr. Stephanie Gil
    • Topics in Reinforcement Learning - ASU Spring 2019 taught by Dr. Dimitri Bertsekas
    • Introduction to Artificial Intelligence - ASU Spring 2019 taught by Dr. Stephanie Gil
    • Planning and Learning Methods in AI - ASU Fall 2018 taught by Dr. Stephanie Gil
    • Principles of Programming - ASU Fall 2018 taught by Rameen Khaliqi
    • Embedded Systems Lab - IIT Bombay Spring, 2014 taught by Dr. Kavi Arya
    • Database and Information Systems Lab - IIT Bombay Autumn, 2014 taught by Dr. N.L. Sarda