Multitask Q-learning in Atari Games This is an implementation of multitask deep Q-learning for Atari games with PyTorch. See here for more details.