Joey Hejna

imitation learning (4) offline reinforcement learning (2) robot manipulation (2) sequential decision making (1) direct preference optimization (1) supervised learning (1) policy learning (1) mutual information (1) generalization ability (1) offline learning (1) temporal abstraction (1) k-nearest neighbor (1) variational autoencoder (1) reward hacking (1) language model (1) reward model (1) policy training (1) goal-conditioned policy (1) preference-based reinforcement learning (1) long-horizon planning (1)