Pranav Ramanathan

I'm Pranav Ramanathan, a Mathematics undergraduate at Queen Mary University of London working on reinforcement learning, large language models, and scientific machine learning. Through the STRIDE Research Programme, I explored AI approaches to the No-Three-in-Line problem using transformers and RL agents, and I am now investigating a physics-inspired UDE formulation of its constraint landscape. I am currently exploring the HP protein-folding model, comparing reinforcement-learning policies with CP-SAT for ground-state search and examining where symbolic and learning-based approaches perform best. This site highlights my research and projects I've worked on.