Sim-to-Lab-to-Real: Safe Reinforcement Learning with Shielding and Generalization Guarantees

Open in new window