Alchemy: A structured task distribution for meta-reinforcement learning

Open in new window