Meraj Mammadov

I am a PhD student in computer science at Örebro University and a WASP fellow, working with Johannes Andreas Stork and Pedro Zuidberg Dos Martires.

My research focuses on reinforcement learning and transfer learning. I study how knowledge from existing data and foundation models can help agents learn new tasks.

Previously, I worked on vision-based navigation for UAVs with Hyondong Oh at UNIST, where I received my bachelor's and master's degrees in mechanical engineering. Earlier, I represented Azerbaijan at international Physics Olympiads and received several national and international awards.

Meraj Mammadov

Publications

  1. RAIL: Representation-Aligned Imitation Learning for Student-Compatible Teacher Policies 2026
    M. Mammadov, P. Zuidberg Dos Martires, and J. A. Stork. Advances in Neural Information Processing Systems (NeurIPS).
  2. Teacher–Student Representational Alignment for Reinforcement Learning-Driven Imitation Learning 2026
    M. Mammadov, P. Zuidberg Dos Martires, and J. A. Stork. ICRA Workshop on Reinforcement Learning for Imitation Learning (RL4IL). Oral presentation.
  3. PACMAN: Rapid Identification of Keypoint Patch-Based Fiducial Markers in Occluded Environments 2025
    T. Park, G. Bae, W. Shin, M. Mammadov, J. Seo, H. Shin, and H. Oh. Image and Vision Computing.
  4. Autonomous Landing on a Moving Platform Using Vision-Based Deep Reinforcement Learning 2024
    P. Ladosz, M. Mammadov, H. Shin, W. Shin, and H. Oh. IEEE Robotics and Automation Letters (RA-L). Presented at IROS.
  5. End-to-End LiDAR-Driven Reinforcement Learning for Autonomous Racing 2024
    M. Mammadov and H. Oh. Korea Robotics Conference (KRoC).