Konrad Zolna

offline reinforcement learning (3) batch reinforcement learning (2) robotic manipulation (1) policy learning (1) value function approximation (1) offline learning (1) reward learning (1) robot manipulation (1) continuous action (1) discrete action (1) visual feature (1) generative adversarial imitation learning (1) state distribution (1) discriminator network (1) discriminator optimization (1) reward sketching (1) task-relevant feature (1) critic regularized regression (1) policy optimization (1)