使用扩散策略引导发现新行为
Guided Discovery of New Behaviors using Diffusion Policies
AI总结：
提出结合Feynman-Kac校正器与引导势能的框架，从扩散策略中挖掘并优化罕见但可行的轨迹，再训练策略以系统发现多样化可执行行为。
摘要：
扩散模型已成为机器人学中生成建模的强大工具，扩散策略在多模态动作-轨迹分布建模方面表现出色。然而，当演示数据有限时，标准采样通常再现主导行为，而忽略有效但罕见的模式，限制了新解决方案的发现。现有方法（如引导方法或将强化学习与扩散结合）要么将样本推入不可行区域，要么难以逃离局部最小值，无法系统地发现多样化行为。为解决这些挑战，我们提出一个框架，将Feynman-Kac校正器与一种新颖的引导势能相结合，系统地将扩散策略样本引导至有前景但代表性不足的样本。这些轨迹通过基于采样的轨迹优化进行精炼，并重新纳入训练集以重新训练扩散策略。我们的方法有效地挖掘和修复新轨迹，实现多样化且可执行行为的系统发现。我们在多种操作环境中展示了该框架的有效性，一致地发现了新行为。
英文摘要：
Diffusion models have become a powerful tool for generative modeling in robotics, with diffusion policies excelling at modeling multimodal action-trajectory distributions. However, when demonstrations are limited, standard sampling often reproduces dominant behaviors while neglecting valid but rare modes, limiting the discovery of novel solutions. Existing approaches, such as guidance methods or combining reinforcement learning with diffusion, either push samples into infeasible regions or struggle to escape local minima, failing to systematically uncover diverse behaviors. To address these challenges, we propose a framework that combines Feynman-Kac correctors with a novel guiding potential that systematically guides diffusion policy samples towards promising yet underrepresented samples. These trajectories are refined using sampling-based trajectory optimization and reincorporated into the training set to retrain the diffusion policy. Our method effectively mines and repairs novel trajectories, enabling the systematic discovery of diverse and executable behaviors. We demonstrate the effectiveness of our framework across a range of manipulation environments, consistently discovering new behaviors.
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。
