World Models

Deep SPI
Source code for replicating the experiments presented in our paper on safe policy improvement via world models
Florent Delgrange, Raphaël Avalos, Willem Röpke
Deep SPI
VenueALA Integrating RL and Planning through Optimal Transport World Models
We propose learning a bisimilar model of the environment through optimal transport and unify this with reinforcement learning and planning.
Willem Röpke, Raphaël Avalos, Roxana Rădulescu, Ann Nowé, Diederik M Roijers, Florent Delgrange