| 发明名称 |
Log-linear dialog manager that determines expected rewards and uses hidden states and actions |
| 摘要 |
A dialog manager receives previous user actions and previous observations and current observations. Previous and current user states, previous user actions, current user actions, future system actions, and future observations are hypothesized. The user states, the user actions, and the user observations are hidden. A feature vector is extracted based on the user states, the system actions, the user actions, and the observations. An expected reward of each current action is based on a log-linear model using the feature vectors. Then, the current action that has an optimal expected reward is outputted. |
| 申请公布号 |
US9311430(B2) |
申请公布日期 |
2016.04.12 |
| 申请号 |
US201314106968 |
申请日期 |
2013.12.16 |
| 申请人 |
Mitsubishi Electric Research Laboratories, Inc. |
发明人 |
Watanabe Shinji;Tang Hao |
| 分类号 |
G06F15/18;G06F17/30;G06F17/27;G10L15/22;G10L15/18 |
主分类号 |
G06F15/18 |
| 代理机构 |
|
代理人 |
Vinokur Gene;Brinkman Dirk |
| 主权项 |
1. A dialog manager comprising the steps of:
receiving previous user actions and previous observations and current observations; hypothesizing previous and current user states, previous user actions, current user actions, future system actions, and future observations, wherein the user states, the user actions, and the user observations are hidden; extracting a feature vector based on the user states, the system actions, the user actions, and the observations; determining an expected reward of each current action based on a log-linear model using the feature vectors; and outputting the current action that has an optimal expected reward, wherein the steps are performed in a processor. |
| 地址 |
Cambridge MA US |
