About Me

I’m a PhD student at NTU, working in the HESL laboratory under the supervision of Assoc. Prof. Arvind Easwaran. My research is focused on machine learning safety and probabilistic guarantees for reinforcement learning using the probably approximately correct (PAC) framework of learning. Over the course of my work, I’ve become familiar with stochastic optimization and its use in risk determination for application in safety-critical settings. Some of the research problems I’ve worked on include improving theoretical sample-complexity bounds for RL as well as developing error-confidence safety assurances for RL agents during deployment.

I am currently investigating probabilistic guarantees within adversarial multi-agent systems.

My Interests

Outside of the lab, I enjoy playing strategy games and I have a fondness for applying learning-based techniques to automate and discover solution strategies to game environments (Tetris, Terminal, Bloons, Minecraft…). I also volunteer with local community service organizations in my free time.

Feel free to reach out to me via any of my socials if you’d like to ping me about research ideas, collaborations or clarifications.

Computing subgoal decision in complex environments Like MineCraft based on semantics

RL agent solutions to Tetris and LunarLander

Recent Updates