Md Mahmudul Hasan Rabbi is an Undergraduate Student Researcher at AVIS Lab and an undergraduate honours student in the Department of Applied Mathematics at the University of Dhaka. He works under the joint supervision of Dr. Md Mehedi Hasan and Prof. Dr. AKM Azad.
His research focuses on mathematical optimization, continuous-time generative models, reinforcement learning for reasoning, and multimodal foundation models.
He has co-authored two research manuscripts submitted to IEEE Transactions on Neural Networks and Learning Systems: SPARK: A Stochastic Optimizer for Fast Convergence in Neural Network and AGL Loss: Optimizing Binary Classification by Asymmetric Generalized Logarithmic Loss Function with Statistical Validation.
His technical work includes implementing deep learning systems in PyTorch, reinforcement learning with Group Relative Policy Optimization (GRPO), non-neural visual tokenization, diffusion models, and Physics-Informed Neural Networks (PINNs).
Research Interests: Deep Learning, Computer Vision, Generative AI, Reinforcement Learning, Mathematical Optimization, Stochastic Modeling, and Medical AI.
Search for Md Mahmudul Hasan Rabbi's papers on the Publications page