Ashvin Nair

goal-conditioned reinforcement learning (2) self-supervised learning (2) representation learning (2) offline reinforcement learning (1) robotic manipulation (1) robot learning (1) off-policy learning (1) robot manipulation (1) skill learning (1) goal-conditioned policy (1) reward shaping (1) skill generalization (1) subgoal planning (1) self-supervised reinforcement learning (1) goal-reaching policy (1) state coverage (1) maximum-entropy goal distribution (1) skill repertoire (1) visual input (1) manipulation task (1)