Hossein Abdi

I am a final-year Ph.D. researcher at The University of Manchester, in the Department of Computer Science (Machine Learning and Robotics group). My research focuses on developing Kalman-based second-order policy optimization algorithms in reinforcement learning. My work provides theoretical convergence guarantees, with large-scale training on GPU clusters using PyTorch/JAX, and deployment on a Franka Panda robotic arm. I am particularly interested in decision-making and control using large-scale vision-language-action models.