Arpan Dasgupta

profile_pic.jpg

Hi, I am Arpan, a Pre-Doctoral Researcher at Google DeepMind. I work on multi-agent systems, RL and decision making. I am on the outlook for PhD positions in the field of Reinforcement Learning.

I currently work in the team Multi-agent Systems for Social Impact, developing bandit algorithms for social-good applications in a team lead by Prof. Milind Tambe (Harvard University) and Dr. Aparna Taneja. Previously, I have worked at MDSR Lab, Adobe on Explainibility on Explainability in Offline RL.

I graduated in 2023 from International Institute of Information Technology (IIIT) Hyderabad with a dual degree of B.Tech (with Honors) and MS by Research where I earned the Programme Gold medal for the highest CGPA. My Masters research was focused on Extreme Classification and Monte Carlo Tree Search based methods under supervision of Prof. Pawan Kumar.

Outside of work, I am highly interested in all kinds of sports (especially football/soccer and TT). I also learn vocals in the Hindustani classical style and am interested in multiple genres of music.

selected publications

  1. Explaining RL Decisions with Trajectories
    Shripad Vilasrao Deshmukh, Arpan Dasgupta, Balaji Krishnamurthy, Nan Jiang, Chirag Agarwal, Georgios Theocharous, and Jayakumar Subramanian
    In The Eleventh International Conference on Learning Representations (ICLR), 2023
  2. Bayesian Collaborative Bandits with Thompson Sampling for Improved Outreach in Maternal Health Program
    Arpan Dasgupta, Gagan Jain, Arun Suggala, Karthikeyan Shanmugam, Milind Tambe, and Aparna Taneja
    AAMAS 2025 (Yet to appear), 2025
  3. Alpha Elimination: Using Deep Reinforcement Learning to Reduce Fill-In During Sparse Matrix Decomposition
    Arpan Dasgupta, and Pawan Kumar
    In Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 2023
  4. Preliminary Study of the Impact of AI-Based Interventions on Health and Behavioral Outcomes in Maternal Health Programs
    Arpan Dasgupta, Niclas Boehmer, Neha Madhiwalla, Aparna Hedge, Bryan Wilder, Milind Tambe, and Aparna Taneja
    AASG, Autonomous Agents and Multi-Agent Systems at AAMAS, 2024