Adam Gleave, Oliver Habryka: Multi-task Maximum Entropy Inverse Reinforcement Learning. CoRR abs/1805.08882 (2018)