Reinforcement Learning for Spoken Dialogue Systems
Reinforcement Learning for Spoken Dialogue Systems
复制标题
口语对话系统的强化学习
DOI：
--
复制
发表时间：
1999
期刊：
影响因子：
--
通讯作者：
M. Walker
中科院分区：
文献类型：
--
作者：
Satinder Singh;Michael Kearns;D. Litman;M. Walker
最近，一些作者提出将对话系统视为马尔可夫决策过程(MDP)。然而，MDP算法在对话系统中的实际应用面临着许多严峻的技术挑战。我们在MDP框架的基础上构建了一个通用的软件工具(RLDs，用于对话系统的强化学习)，并将其应用于从AT&T实验室建立的两个对话系统中收集的对话语料库。我们的实验表明，RLD有望成为“浏览”和理解复杂的、时间相关的对话语料库中的相关性的工具。
Recently, a number of authors have proposed treating dialogue systems as Markov decision processes (MDPs). However, the practical application of MDP algorithms to dialogue systems faces a number of severe technical challenges. We have built a general software tool (RLDS, for Reinforcement Learning for Dialogue Systems) based on the MDP framework, and have applied it to dialogue corpora gathered from two dialogue systems built at AT&T Labs. Our experiments demonstrate that RLDS holds promise as a tool for "browsing" and understanding correlations in complex, temporally dependent dialogue corpora.