Reinforcement Learning (RL) is a data-driven framework for modeling and solving sequential decision-making problems in uncertain environments.

After a general introduction to the key ideas behind RL, such as learning through interaction, reward optimization, and the balance between exploration and exploitation, we will explore two recent applications that highlight the flexibility of this approach in different domains.

The first case study addresses decision-making in a medical context, where RL techniques can support personalized strategies over time. The second focuses on autonomous systems and the role of learning in environments with limited or partial information.

Both examples emphasize how uncertainty and temporal structure can be naturally integrated within the RL framework, offering insights into its practical implementation and methodological challenges.

For online partecipation, please register below.