My research interests broadly include reinforcement learning, theoretical machine learning and the alignment problem of AGI. My Bachelor’s thesis is under Prof. Shivaram Kalyanakrishnan in which I worked on building an AI agent for the game of Reconnaissance Blind Chess, which involves planning under partial observability. Our agent, Fianchetto won the NeurIPS 2021 Competition on RBC.
I have also worked with Prof. Ganesh Ramakrishnan in collaboration with Adobe Research India on the problem of understanding structured documents with pretrained models. In the summer of 2020, I interned at the PLRI lab of Technical University of Braunschweig, Germany to work with Prof. Thomas Deserno.
In my spare time, I enjoy playing chess, watching movies and listening to music.
|Jan 31, 2022||Accepted for the Research Week with Google in the ML Foundations sub-track|
|Jan 15, 2022||Accepted for the AI Alignment track in the AGI Safety Fundamentals Programme conducted by Effective Altruism Cambridge|
|Dec 8, 2021||Presented the algorithm for my AI bot Fianchetto, at the NeurIPS 2021 workshop on RBC|
|Oct 23, 2021||Won the NeurIPS 2021 Competition on Reconnaissance Blind Chess (RBC)|
The Second NeurIPS Tournament of Reconnaissance Blind ChessIn Proceedings of the NeurIPS 2021 Competitions and Demonstrations Track 06–14 dec 2022
Multi-camera, multi-person, and real-time fall detection using long short term memoryIn Medical Imaging 2021: Imaging Informatics for Healthcare, Research, and Applications 2021
Fianchetto: Speed, Belief, Guile, Caution to Win at Reconnaissance Blind Chesssubmitting to The Journal of Artificial Intelligence Research 2022
Randomized POMDP Planning AlgorithmsTechnical Report, 2021