|
| My research concerns the design of intelligent computational
systems capable to work autonomously in their environments, in order to
accomplish goals safety and accurately. I am particularly interested in
those computational systems, also called agents, that are able to
improve their performances over time by means of supervised or
reinforcement learning. At the present time, I am investigating
architectures and algorithms for agents that benefit from the
information provided by advisors in the form of behavior models, and
from the experience generated by themselves through the interaction
with their environments. My current research focuses mainly on the
combination of function approximators and reinforcement learning
algorithms based on temporal differences. |
