I am Samuele Tosatto, an Assistant Professor at the University of Innsbruck. I was previously a postdoc in the RLAI Lab, specifically in the R-LAIR headed by Prof. Mahmood. I took my Ph.D. under the supervision of Prof. Jan Peters at the Intelligent Autonomous System lab (IAS), defending my Ph.D. dissertation “Towards Off-Policy Reinforcement Learning for Robotics”.
My current research goal is to develop embodied learning agents.
Embodied agents are real-world systems, like robots, that perceive and interact with the real world. Most of the current autonomous systems have limited intelligence and understanding of the world. I am interested in making those systems more intelligent and able to learn by interacting with the real world.
Most current research focuses on training agents in a simulated environment and then transferring such knowledge into the real world.
While this approach simplifies many challenges present in the real world, it is hardly scalable as it requires the design of new virtual environments for different tasks! Furthermore, I believe that the real challenge is not to develop pre-trained robots but to have robots that continuously learn in the world, just like we do!
To solve this challenge, I focus my research on reinforcement learning (a technique that allows agents to learn from a reward signal), representation learning for robotics, (e.g., movement primitives), and abstraction.
Office: Technicherstraße 21, 3N07.
Contact me via email: samuele.tosatto@uibk.ac.at samuele.tosatto@gmail.com.