arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1606.02689cs.CLcs.LG
收藏

持续学习神经对话管理

Continuously Learning Neural Dialogue Management

University of Cambridge(剑桥大学)

Pei-Hao Su, Milica Gasic, Nikola Mrksic, Lina Rojas-Barahona, Stefan Ultes, David Vandyke, Tsung-Hsien Wen, Steve Young

摘要:

我们描述了一个两步的方法,在面向任务的口语对话系统的对话管理。提出了一个统一的神经网络框架,使系统能够首先通过监督从一组对话数据中学习,然后通过强化学习不断改善其行为,所有这些都使用基于梯度的算法在一个单一的模型上。实验表明,监督模型的有效性,在基于语料库的评价,与用户模拟,并与付费人类科目。强化学习的使用进一步提高了模型在两种交互式环境中的性能,特别是在高噪声条件下。

英文摘要:

We describe a two-step approach for dialogue management in task-oriented spoken dialogue systems. A unified neural network framework is proposed to enable the system to first learn by supervision from a set of dialogue data and then continuously improve its behaviour via reinforcement learning, all using gradient-based algorithms on one single model. The experiments demonstrate the supervised model's effectiveness in the corpus-based evaluation, with user simulation, and with paid human subjects. The use of reinforcement learning further improves the model's performance in both interactive settings, especially under higher-noise conditions.

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。

↑