arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.13703cs.ROcs.LGstat.AP
收藏

面向机器人学习的带统计保证的任务驱动型分布偏移检测

Task-Driven Detection of Distribution Shifts with Statistical Guarantees for Robot Learning

Princeton University(普林斯顿大学);NVIDIA(英伟达)

Alec Farid, Sushant Veer, Divyanshu Pachisia, Anirudha Majumdar

AI总结:

该研究针对机器人学习中的分布偏移检测问题,基于PAC-贝叶斯理论与统计技术提出任务驱动的OOD检测方法,具备假阳性和假阴性率的统计保证,少量试验即可在抓取、避障任务中生效。

摘要:

我们的目标是执行分布外(OOD)检测,即检测机器人何时在与训练机器人所用环境分布不同的环境中运行。我们利用概率近似正确(PAC)-贝叶斯理论来训练策略,该策略在训练分布上的性能具有保证界。我们的OOD检测思路基于以下直觉:测试环境中性能界的违反可作为机器人处于分布外运行的证据。我们通过基于p值和集中不等式的统计技术将这一思路形式化。该方法为OOD检测提供有保证的置信界,包括检测器的假阳性率和假阴性率界,且是任务驱动的,仅对影响机器人性能的变化敏感。我们在仿真和硬件上验证了该方法,实验包括使用形状或位姿不熟悉的物体的抓取任务,以及无人机在有风扰和不同障碍物密度环境中执行基于视觉的避障任务。我们的示例表明,仅需少量试验即可实现任务驱动的OOD检测。

英文摘要:

Our goal is to perform out-of-distribution (OOD) detection, i.e., to detect when a robot is operating in environments drawn from a different distribution than the ones used to train the robot. We leverage Probably Approximately Correct (PAC)-Bayes theory to train a policy with a guaranteed bound on performance on the training distribution. Our idea for OOD detection relies on the following intuition: violation of the performance bound on test environments provides evidence that the robot is operating OOD. We formalize this via statistical techniques based on p-values and concentration inequalities. The approach provides guaranteed confidence bounds on OOD detection including bounds on both the false positive and false negative rates of the detector and is task-driven and only sensitive to changes that impact the robot's performance. We demonstrate our approach in simulation and hardware for a grasping task using objects with unfamiliar shapes or poses and a drone performing vision-based obstacle avoidance in environments with wind disturbances and varied obstacle densities. Our examples demonstrate that we can perform task-driven OOD detection within just a handful of trials.

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。

↑