Horaire

09:30 - 10:00 Welcome and coffee
10:00 - 10:40 Red-Teaming for Robustness - Niki Howe
10:40 - 11:20 A Distributional Analogue to the Successor Representation - Harley Wiltzer
11:20 - 11:50 Offline constraint inference for safe reinforcement learning - Adriana Hugessen
12:00 - 13:00 Lunch
13:00 - 13:40 Regularization and Robustness in Reinforcement Learning - Esther Derman
13:40 - 14:20 On Convex Optimization Formulation of Robust MDPs - Marek Petrik
14:20 - 14:50 Coffee Break
14:50 - 15:30 Deep Risk Averse Reinforcement Learning under Dynamic Expectile Risk Measures - Erick Delage
15:30 - 16:10 Dynamic Program Decomposition of Static Risk Measures - Jia Lin Hau
16:10 - 16:50 Risk-aware Restless Bandits: Safe-Whittle Index - Nima Akbarzadeh