How to pick the domain randomization parameters for sim-to-real transfer of reinforcement learning policies?

Open in new window