Amir Mohammad Karimi Mamaghan

I’m a WASP and ELLIS Ph.D. student at the Department of Decision and Control Systems (DCS) of KTH Royal Institute of Technology under the supervision of Prof. Karl Henrik Johansson and Prof. Stefan Bauer. I completed my master’s degree in Computer Engineering (Artificial Intelligence and Robotics) at the University of Tehran. During my master’s studies, I was a research assistant at Data Analytics Lab under the supervision of Prof. Behnam Bahrak.

Research Interests

My research interests lie at the intersection of representation learning, causality, and foundation models. I focus on how high-level structure emerges in learned representations, such as causal variables and object-level abstractions, and how to evaluate these representations beyond i.i.d. performance, especially under uncertainty and distribution shifts. More recently, I am investigating training dynamics in deep transformer models. Overall, my goal is to better understand and steer representation learning to build models that are more robust, interpretable, and capable of systematic generalization across domains.