John Schulman

deep reinforcement learning (5) reinforcement learning (5) policy gradient (4) sample efficiency (3) reward model (3) data augmentation (2) reinforcement learning from human feedback (2) exploration bonus (2) intrinsic motivation (2) neural network (2) continuous control (2) instruction following (1) image generation (1) language model alignment (1) motion planning (1) variational inference (1) model alignment (1) domain generalization (1) gradient estimation (1) policy optimization (1)