Papers
14
Total Citations
204
H-Index
7
About
Nikolai Matni is a leading figure at the intersection of control theory, reinforcement learning, and robotics, whose work is shaping how autonomous systems learn and operate safely in the physical world. His research focuses on developing rigorous, data-driven frameworks that bridge the gap between classical control and modern machine learning. Matni’s most influential work, “From self-tuning regulators to reinforcement learning and back again” (with over 90 combined citations), provides a foundational perspective on unifying adaptive control and RL. He has made seminal contributions to learning-based control with formal guarantees, including pioneering methods for learning stability certificates directly from data and analyzing the sample complexity of imitation learning for stable systems. His recent work on a quantitative framework for layered multirate control (2024) is establishing a new theory of control architecture for complex systems. Matni’s impact extends to practical robotics, with notable achievements in vision-based quadrotor obstacle avoidance using transformers and safe motion planning with temporal logic. His research is highly cited and influential, with his top papers accumulating over 200 citations, reflecting his role as a key architect of the next generation of safe, learning-enabled autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1From self-tuning regulators to reinforcement learning and back again74 citations · 2019
- 2
- 3Learning Stability Certificates from Data28 citations · 2020
- 4On the Sample Complexity of Stability Constrained Imitation Learning17 citations · 2021
- 5From self-tuning regulators to reinforcement learning and back again16 citations · 2019
- 6
- 7
- 8
- 9
- 10