DrS: Learning Reusable Dense Rewards for Multi-Stage Tasks

Documenti analoghi

• Policy Decorator: Model-Agnostic Online Refinement for Large Policy Model
  di: Yuan, Xiu, et al.
  Pubblicazione: (2024)
• Multi-Stage Manipulation with Demonstration-Augmented Reward, Policy, and World Model Learning
  di: Escoriza, Adrià López, et al.
  Pubblicazione: (2025)
• TimeRewarder: Learning Dense Reward from Passive Videos via Frame-wise Temporal Distance
  di: Liu, Yuyang, et al.
  Pubblicazione: (2025)
• Towards Embodiment Scaling Laws in Robot Locomotion
  di: Ai, Bo, et al.
  Pubblicazione: (2025)
• When Should We Prefer State-to-Visual DAgger Over Visual Reinforcement Learning?
  di: Mu, Tongzhou, et al.
  Pubblicazione: (2024)