Fetching the paper…

Situated GAIL: Multitask imitation using task-conditioned adversarial inverse reinforcement learning · Around