Reinforcement Learning for Stochastic Networks, Toulouse

Session

Keynote: Vincent François-Lavet (VU, Amsterdam)

Jun 20, 2024, 9:30 AM

Amphi B00 (ENSEEIHT)

78. Generalization in deep reinforcement learning and the role of representation learning for sequential decision-making tasks

6/20/24, 9:30 AM

Abstract: The concept of generalization in the context of deep reinforcement learning will be explained along with the different elements that can be used to improve generalization. The key role of representation learning for improved generalization and interpretability will then be shown. Finally, the role of representation learning for exploration and transfer learning will also be discussed.