MiniGrid

Dataset generated from the MiniGrid-FourRooms environment. The objective of the agent is to reach a goal position in a gridworld. We regenerate the dataset of D4RL for full reproducibility, using a random policy and an expert policy that navigates straight to the goal.

References

[1] Fu, Justin, et al. ‘D4RL: Datasets for Deep Data-Driven Reinforcement Learning’. CoRR, vol. abs/2004.07219, 2020, https://arxiv.org/abs/2004.07219.