Video Prediction Models as Rewards for Reinforcement Learning Using advances in generative modeling to learn reward functions from unlabeled videos. Barkour: Benchmarking animal-level agility with quadruped robots We introduce the Barkour agility benchmark for quadruped robots, along with a Transformer-based generalist locomotion policy DayDreamer: World Models for Physical Robot Learning Hyper data-efficient robot learning with world models Learning a Diffusion Model Policy from Rewards via Q-Score Matching Learning diffusion model policies for off-policy reinforcement learning. Adversarial Motion Priors Make Good Substitutes for Complex Reward Functions Using motion priors to learn energy-efficient and natural robot locomotion strategies Learning to Walk on Complex Terrains with Vision Examining the role of vision in robot locomotion Autonomously Untangling Long Cables Sliding and Grasping for Tangle Manipulation (SGTM)
