arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.11729cs.CLcs.AI
收藏

GoChat:基于分层强化学习的目标导向聊天机器人

GoChat: Goal-oriented Chatbots with Hierarchical Reinforcement Learning

Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所);UCAS(中国科学院大学)

Jianfeng Liu, Feiyang Pan, Ling Luo

摘要:

一个像人类一样交谈的聊天机器人应该是目标导向的(即,在对话中有目的),这超出了语言生成。然而,现有的对话系统通常严重依赖于繁琐的手工制作的规则或昂贵的标记数据集来实现目标。在本文中,我们提出了面向目标的聊天机器人(GoChat),这是一个端到端训练聊天机器人的框架,可以最大限度地提高离线多轮对话数据集的长期回报。我们的框架利用分层强化学习(HRL),其中高层政策通过确定一些子目标来引导会话朝向最终目标,而低层政策通过生成相应的话语来响应来实现子目标。在我们对真实世界的金融反欺诈对话数据集的实验中,我们的方法在响应生成的质量和完成目标的成功率方面都优于以前的方法。

英文摘要:

A chatbot that converses like a human should be goal-oriented (i.e., be purposeful in conversation), which is beyond language generation. However, existing dialogue systems often heavily rely on cumbersome hand-crafted rules or costly labelled datasets to reach the goals. In this paper, we propose Goal-oriented Chatbots (GoChat), a framework for end-to-end training chatbots to maximize the longterm return from offline multi-turn dialogue datasets. Our framework utilizes hierarchical reinforcement learning (HRL), where the high-level policy guides the conversation towards the final goal by determining some sub-goals, and the low-level policy fulfills the sub-goals by generating the corresponding utterance for response. In our experiments on a real-world dialogue dataset for anti-fraud in financial, our approach outperforms previous methods on both the quality of response generation as well as the success rate of accomplishing the goal.

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。

↑