About
Appointed Canada CIFAR AI Chair – 2026
Jun Jin investigates how intelligent machines can learn and adapt continuously through real-world interactions. He focuses on the intersection of reinforcement learning (RL) and embodied artificial intelligence (EAI), studying how robots acquire, refine, and reuse motor skills through direct interaction with their physical environment.
As AI systems move into open-ended real-world environments, a key challenge is enabling learning that scales through experience rather than data or model size alone. He is interested in developing computational models that enable robots to build predictive internal representations of their own actions and outcomes, rather than merely static predictions, enabling lifelong learning agents that operate robustly across changing tasks, environments, and human needs. This direction supports the development of robots as general-purpose tools — systems that learn over time, collaborate with people, and remain adaptable in big world settings – an ideal form of “human-centered autonomy”, which is his vision for AI/Robotics research.
Awards
- ICRA 2022 Outstanding Student Paper Finalist (top 3 runner-up), IEEE Robotics and Automation Society (2022).
- Alberta Graduate Excellence Scholarship, Government of Alberta (2020)
- KUKA Innovation Award Global Top 5 Finalist, KUKA Robotics (2018)
Relevant Publications
- Jin, J., Graves, D., Haigh, C., Luo, J., & Jagersand, M. (2022, May). Offline learning of counterfactual predictions for real-world robotic reinforcement learning. In 2022 International Conference on Robotics and Automation (ICRA) (pp. 3616-3623). IEEE.
- Jin, J., Nguyen, N. M., Sakib, N., Graves, D., Yao, H., & Jagersand, M. (2020, May). Mapless navigation among dynamics with social-safety-awareness: a reinforcement learning approach from 2d laser scans. In 2020 IEEE international conference on robotics and automation (ICRA) (pp. 6979-6985). IEEE.
- Jin, J., Zhang, H., & Luo, J. Build generally reusable agent-environment interaction models. In NeurIPS 2022 Foundation Models for Decision Making Workshop.