Adversarial Soft Advantage Fitting: Imitation Learning Without RL

1 points| gbachik | 5 years ago |arxiv.org