arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2011.02511cs.CLcs.LG
收藏

基于人的反馈的离线强化学习

Offline Reinforcement Learning from Human Feedback in Real-World Sequence-to-Sequence Tasks

Google Research(谷歌研究院);Heidelberg University(海德堡大学);NEC Laboratories Europe(NEC欧洲实验室)

Julia Kreutzer, Stefan Riezler, Carolin Lawrence

摘要:

可以从部署在现实世界中的NLP系统中收集大量的交互日志。如何利用这些丰富的信息?在离线强化学习(RL)设置中使用这样的交互日志是一种有前途的方法。然而,由于NLP任务的性质和生产系统的限制,出现了一系列的挑战。我们对这些挑战进行了简要概述,并讨论了可能的解决方案。

英文摘要:

Large volumes of interaction logs can be collected from NLP systems that are deployed in the real world. How can this wealth of information be leveraged? Using such interaction logs in an offline reinforcement learning (RL) setting is a promising approach. However, due to the nature of NLP tasks and the constraints of production systems, a series of challenges arise. We present a concise overview of these challenges and discuss possible solutions.

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。

↑