arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2110.08258cs.LGcs.AIcs.HCcs.RO
收藏

一个学习从人类那里请求丰富和上下文有用信息的框架

A Framework for Learning to Request Rich and Contextually Useful Information from Humans

University of Maryland, College Park(马里兰大学帕克分校);Carnegie Mellon University(卡内基梅隆大学);Microsoft Research(微软研究院)

Khanh Nguyen, Yonatan Bisk, Hal Daumé

摘要:

当部署时,人工智能智能体将遇到超出其自主解决问题能力的问题。利用人类的帮助可以帮助智能体克服其固有的局限性,并有力地应对不熟悉的情况。我们提出了一个通用的交互式框架,使代理请求和解释丰富的,上下文有用的信息,从一个助手,有知识的任务和环境。我们证明了我们的框架的实用性,在一个模拟的人类辅助导航问题。借助我们的方法学习的援助请求策略,与完全自主的行为相比,导航代理在以前看不见的环境中执行任务的成功率提高了7倍。我们表明,代理可以根据上下文利用不同类型的信息,并分析学习的援助请求政策时,助理可以递归分解成子任务的好处和挑战。

英文摘要:

When deployed, AI agents will encounter problems that are beyond their autonomous problem-solving capabilities. Leveraging human assistance can help agents overcome their inherent limitations and robustly cope with unfamiliar situations. We present a general interactive framework that enables an agent to request and interpret rich, contextually useful information from an assistant that has knowledge about the task and the environment. We demonstrate the practicality of our framework on a simulated human-assisted navigation problem. Aided with an assistance-requesting policy learned by our method, a navigation agent achieves up to a 7x improvement in success rate on tasks that take place in previously unseen environments, compared to fully autonomous behavior. We show that the agent can take advantage of different types of information depending on the context, and analyze the benefits and challenges of learning the assistance-requesting policy when the assistant can recursively decompose tasks into subtasks.

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。

↑