Nature
Exploring the limits of hierarchical world models in reinforcement learning - Scientific Reports
Specifically, we present the inner workings of our hierarchical world model, the agents that are trained via word model simulations and the process we use to temporally abstract the
1 day ago