ReWiND：语言引导的奖励在无需新演示的情况下教会机器人策略
ReWiND: Language-Guided Rewards Teach Robot Policies without New Demonstrations
AI总结：
ReWiND 框架利用语言指令和小型演示数据集学习语言条件奖励函数并预训练策略，在无需新演示的情况下实现样本高效的任务适应，显著优于基线。
摘要：
我们提出 ReWiND，一个仅从语言指令学习机器人操作任务而无需逐任务演示的框架。标准强化学习（RL）和模仿学习方法需要专家监督，通过人工设计的奖励函数或每个新任务的演示来实现。相比之下，ReWiND 从一个小的演示数据集开始学习：（1）一个数据高效的、以语言为条件的奖励函数，用奖励标记该数据集；（2）一个以语言为条件的策略，使用这些奖励通过离线 RL 进行预训练。给定一个未见过的任务变体，ReWiND 使用学习到的奖励函数微调预训练策略，只需极少的在线交互。我们表明，ReWiND 的奖励模型能有效泛化到未见过的任务，在奖励泛化和策略对齐指标上比基线高出最多 2.4 倍。最后，我们证明 ReWiND 能够实现样本高效地适应新任务，在仿真中比基线高出 2 倍，并将真实世界预训练双臂策略提升 5 倍，向可扩展的真实世界机器人学习迈出了一步。详见网站 https://rewind-reward.github.io/。
英文摘要：
We introduce ReWiND, a framework for learning robot manipulation tasks solely from language instructions without per-task demonstrations. Standard reinforcement learning (RL) and imitation learning methods require expert supervision through human-designed reward functions or demonstrations for every new task. In contrast, ReWiND starts from a small demonstration dataset to learn: (1) a data-efficient, language-conditioned reward function that labels the dataset with rewards, and (2) a language-conditioned policy pre-trained with offline RL using these rewards. Given an unseen task variation, ReWiND fine-tunes the pre-trained policy using the learned reward function, requiring minimal online interaction. We show that ReWiND's reward model generalizes effectively to unseen tasks, outperforming baselines by up to 2.4x in reward generalization and policy alignment metrics. Finally, we demonstrate that ReWiND enables sample-efficient adaptation to new tasks, beating baselines by 2x in simulation and improving real-world pretrained bimanual policies by 5x, taking a step towards scalable, real-world robot learning. See website at https://rewind-reward.github.io/.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
