Tamara Gerbert
Papers
2
Total Citations
28
H-Index
2
About
Tamara Gerbert is a researcher advancing the frontier of explainable robotics through deep reinforcement learning (DRL). Her work centers on bridging the gap between simulation-trained agents and real-world deployment, with a particular focus on domain randomisation—a technique that varies simulated environments to improve transferability. Gerbert’s most-cited paper, “Analysing deep reinforcement learning agents trained with domain randomisation” (2022, 21 citations), critically examines the trade-off between the robustness of DRL agents and their lack of interpretability compared to classical control methods. In her earlier 2019 study (7 citations), she laid foundational insights into training robots in simulation for real-world tasks. By systematically analyzing agent behavior under domain randomisation, Gerbert has contributed to making DRL more transparent and reliable for practical robotics applications. Her work is especially valuable for students and researchers seeking to understand not just how to train effective agents, but how to trust and interpret their decisions in safety-critical environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2