Parham Mohammad Panahi

I am a machine learning and reinforcement learning researcher and incoming PhD student at the University of Alberta, working with Adam White. I am also affiliated with the RLAI lab and the Alberta Machine Intelligence Institute (amii). Before that, I graduated with a M.Sc. in Computing Science from the University of Alberta in 2024.

Email  /  CV  /  Scholar  /  Github  /  M.Sc. Thesis

Research

Keywords: Machine Learning, Model-Based Reinforcement Learning, Representation Learning, Continual Learning, and Generative Models.

I am broadly interested in machine learning and reinforcement learning. My research goal is to create efficient and reliable learning systems capable of continual interaction and decision-making in a changing world.

We argue and demonstrate the pitfalls of lifetime tuning in continual reinforcement learning, explain why recent progress in continual RL has been mixed, and motivate the development of empirical practices that better match the goals of continual RL.

We present insight into the interaction between prioritization, bootstrapping, and neural networks and propose several improvements for prioritized replay in tabular settings and noisy domains.

We constrain background planning to a given set of (abstract) subgoals and learning only local, subgoal-conditioned models to avoid compounding model error. Also appeared as an oral presentation at Planning and Reinforcement Learning Workshop at ICAPS 2024.

Recorded Talks

Tea Time Talks 2025, University of Alberta. (Recording available soon!)

Tea Time Talks 2024, University of Alberta.