Filters








1 Hit in 3.2 sec

Deterministic and Discriminative Imitation (D2-Imitation): Revisiting Adversarial Imitation for Sample Efficiency [article]

Mingfei Sun, Sam Devlin, Katja Hofmann, Shimon Whiteson
2022 arXiv   pre-print
We revisit the foundation of adversarial imitation and propose an off-policy sample efficient approach that requires no adversarial training or min-max optimization.  ...  Combined, these insights yield a practical algorithm, Deterministic and Discriminative Imitation (D2-Imitation), which operates by first partitioning samples into two replay buffers and then learning a  ...  In this paper, we revisit the foundation of adversarial imitation and propose an off-policy learning approach, Deterministic and Discriminative Imitation (D2-Imitation), that involves no adversarial training  ... 
arXiv:2112.06054v3 fatcat:zjmy7id6kbbzder67tx4jqesqi