发表机构
TU Munich; Siemens AG（慕尼黑工业大学; 西门子公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对高度依赖接触的操作中数据与学习的问题，提出结合离线深度强化学习的自动数据收集方案，仅用少量自主数据收集，在四个现实任务中平均成功率达96%，为相关任务的数据收集和精密应用高成功率奠定基础。
AI 中文摘要
在大型数据集上进行端到端训练的策略在高精度任务中往往很脆弱，泛化能力也较差。研究发现这些局限性主要源于数据收集缺乏结构和重点。关键见解是仅在高度依赖接触的任务关键部分利用密集数据收集，在简单自由空间运动时依靠传统规划。提出结合离线深度强化学习的自动数据收集方案，消除对远程操作员技能和在线策略更新的依赖。在四个具有挑战性的现实世界任务中，仅用2至2.5小时自主数据收集，平均成功率达96%，而最强基线为55%。在端到端方法难以应对的分布外场景中性能依然很高。研究结果为高度依赖接触的任务的针对性数据收集和精密应用中的高成功率铺平了道路。
英文摘要
Learned policies trained end-to-end on large datasets often remain brittle in high-precision tasks and struggle with generalization. We find that these limitations largely stem from a lack of structure and focus in data collection. Our key insight is to leverage dense data collection only for the critical segment of contact-rich tasks and to rely on traditional planning during simple free-space motion. We propose an automated data-collection scheme in combination with offline deep reinforcement learning for the critical segment of the task, eliminating reliance on a teleoperator's skill and on online policy updates. Across four challenging real-world tasks, using only 2 to 2.5 hours of autonomous data collection, we achieve an average success rate of 96%, compared to the strongest baseline at 55%. Notably, performance remains high in out-of-distribution scenarios where end-to-end approaches struggle. Our results pave the way for targeted data collection for contact-rich tasks and for high success rates in precision applications.
CommentsProject webpage: https://anonymous.4open.science/w/data_and_learning_where_it_matters/
