发表机构
Beijing Institute of Technology; University of Virginia; Shandong University; Northeastern University（北京理工大学; 弗吉尼亚大学; 山东大学; 东北大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对人形机器人强化学习策略中的运动遗忘问题，提出首个运动级遗忘方法ForgetMimic，在保持其他运动性能的同时消除指定运动记忆，并在Unitree G1和H2机器人上验证了有效性。
AI 中文摘要
利用人类演示的人形机器人控制，通过强化学习（RL）已实现了多样、敏捷且自然的运动行为。尽管这一范式在物理人形机器人控制中取得了显著性能，但如何从已学习的策略中消除特定运动仍未得到充分探索。解决这一问题受到紧迫的安全与隐私问题的驱动：移除恶意、中毒或次优运动，以及受版权保护且根据GDPR等法规享有被遗忘权的运动，具有至关重要的意义。为此，我们提出了ForgetMimic，这是首个专为物理世界人形机器人控制设计的运动级遗忘方法。ForgetMimic的核心思想如下：给定一个在N个运动上训练的策略$π_θ$，我们的方法在目标子集K个运动上降低性能，同时保持其余N-K个运动的有效性。此外，我们识别并解决了机器人控制中导致遗忘失败的两个关键训练机制。我们在Unitree G1和H2人形机器人上进行了广泛实验，涵盖12个运动，包括舞蹈、格斗、翻转等。实验结果表明，ForgetMimic有效消除了指定运动的记忆，同时维持所有其他运动的正常运行。
英文摘要
Humanoid control, leveraging human demonstrations, has achieved diverse, agile, and natural locomotion behaviors through reinforcement learning (RL). While this paradigm has yielded remarkable performance in physical humanoid control, how to eliminate specific motions from learned policies remains insufficiently explored. Addressing this issue is motivated by pressing safety and privacy concerns: the removal of malicious, poisoned, or suboptimal motions, as well as copyright-protected motions subject to the right to be forgotten under regulations such as the GDPR, is of critical importance. To this end, we propose {ForgetMimic}, the first motion-level unlearning method designed specifically for physical-world humanoid control. The core idea of ForgetMimic is as follows: given a policy $π_θ$ trained on $N$ motions, our method degrades performance on a target subset of $K$ motions while preserving the effectiveness of the remaining $N-K$ motions. Furthermore, we identify and resolve two key training mechanisms in robot control that lead to unlearning failure. We conduct extensive experiments on the Unitree G1 and H2 humanoid robots across 12 motions, including Dance, Fight, Flip, and others. Experimental results demonstrate that ForgetMimic effectively eliminates memory of designated motions while maintaining the normal operation of all other motions.
Commentshttps://github.com/Zili1000/ForgetMimic
