Timothy Mann
Papers
4
Total Citations
58
H-Index
4
About
Timothy Mann is a leading researcher in reinforcement learning (RL), with a particular focus on making algorithms robust and reliable for real-world control tasks. His most influential work, "Robust Reinforcement Learning for Continuous Control with Model Misspecification" (2019), has garnered 38 citations and introduces a critical framework for handling perturbations in transition dynamics—a common failure point in deployed RL systems. By integrating robustness directly into continuous control algorithms, Mann addresses the gap between simulated training and unpredictable physical environments. Earlier in his career, he explored the intersection of robotics and cognitive development, notably in "Autonomous and Interactive Improvement of Binocular Visual Depth Estimation through Sensorimotor Interaction" (2012), where he demonstrated how humanoid robots can learn depth perception through limited, infant-like interaction. This work highlights his broader interest in biologically inspired learning and sensorimotor feedback. Mann’s contributions are essential for advancing safe, adaptive AI systems, and his research continues to shape how robots learn and operate under uncertainty.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4