Using Wizard-of-Oz simulations to bootstrap Reinforcement - Learning based dialog management systems
Using Wizard-of-Oz simulations to bootstrap Reinforcement - Learning based dialog management systems
复制标题
使用绿野仙踪模拟来引导强化 - 基于学习的对话管理系统
DOI：
--
复制
发表时间：
2003
期刊：
影响因子：
--
通讯作者：
S. Young
中科院分区：
文献类型：
--
作者：
J. Williams;S. Young
本文描述了一种使用Wizard-ofOz试验“引导”基于强化学习的对话管理器的方法。通过注释发现状态空间和动作集，并使用监督学习算法生成初始策略。该方法进行了测试，并显示创建一个初始的政策，执行显着更好，更少的努力比手工制作的政策，并可以使用少量的对话框生成。
This paper describes a method for “bootstrapping” a Reinforcement Learningbased dialog manager using a Wizard-ofOz trial. The state space and action set are discovered through the annotation, and an initial policy is generated using a Supervised Learning algorithm. The method is tested and shown to create an initial policy which performs significantly better and with less effort than a handcrafted policy, and can be generated using a small number of dialogs.