4.5 Article

Expert Intervention Learning An online framework for robot learning from explicit and implicit human feedback

期刊

AUTONOMOUS ROBOTS
卷 46, 期 1, 页码 99-113

出版社

SPRINGER
DOI: 10.1007/s10514-021-10006-9

关键词

-

资金

  1. DARPA Dispersed Computing program, NIH R01 [R01EB019335]
  2. NSF CPS [1544797]
  3. NSF NRI [1637748]
  4. Office of Naval Research
  5. Honda Research Institute USA
  6. RCTA
  7. Amazon

向作者/读者索取更多资源

This study introduces the Expert Intervention Learning (EIL) method, which learns collision avoidance for robots in real and simulated driving tasks through expert interventions. The approach can learn from just a few hundred samples (about one minute) of expert control.
Scalable robot learning from human-robot interaction is critical if robots are to solve a multitude of tasks in the real world. Current approaches to imitation learning suffer from one of two drawbacks. On the one hand, they rely solely on off-policy human demonstration, which in some cases leads to a mismatch in train-test distribution. On the other, they burden the human to label every state the learner visits, rendering it impractical in many applications. We argue that learning interactively from expert interventions enjoys the best of both worlds. Our key insight is that any amount of expert feedback, whether by intervention or non-intervention, provides information about the quality of the current state, the quality of the action, or both. We formalize this as a constraint on the learner's value function, which we can efficiently learn using no regret, online learning techniques. We call our approach Expert Intervention Learning (EIL), and evaluate it on a real and simulated driving task with a human expert, where it learns collision avoidance from scratch with just a few hundred samples (about one minute) of expert control.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.5
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据