Teodor Mihai Moldovan
Papers
2
Total Citations
48
H-Index
2
About
Teodor Mihai Moldovan is a researcher whose work lies at the intersection of robotics, reinforcement learning, and control theory, with a particular focus on model-based approaches. His major contribution is the development of a novel method for model-based reinforcement learning that integrates parametrized physical models with optimism-driven exploration. This approach combines model identification and model predictive control, using a feature-based representation of dynamics that can be fitted with a simple least squares procedure—making it both computationally efficient and practically deployable on robotic platforms. His most cited paper, “Model-based reinforcement learning with parametrized physical models and optimism-driven exploration” (2016), has garnered 45 citations, reflecting its impact on the field. This work is notable for bridging the gap between data-driven learning and physical modeling, enabling robots to learn more efficiently by leveraging prior knowledge of their own dynamics. Moldovan’s contributions are particularly relevant for researchers interested in sample-efficient learning, safe exploration, and the integration of physics-based priors into modern reinforcement learning frameworks.
Research Focus
Key Achievements
Top Papers
- 1
- 2