图 书 馆
缩小差距：最优POMDP的解决方案改进的界限Closing the Gap: Improved Bounds on Optimal POMDP Solutions |
|
| 课程网址: | http://videolectures.net/icaps2011_poupart_solutions/ |
| 主讲教师: | Pascal Poupart |
| 开课单位: | 滑铁卢大学 |
| 开课时间: | 2011-07-21 |
| 课程语种: | 英语 |
| 中文简介: |  |
| 课程简介: | POMDP algorithms have made significant progress in recent years by allowing practitioners to find good solutions to increasingly large problems. Most approaches (including point-based and policy iteration techniques) operate by refining a lower bound of the optimal value function. Several approaches (e.g., HSVI2, SARSOP, grid-based approaches and online forward search) also refine an upper bound. However, approximating the optimal value function by an upper bound is computationally expensive and therefore tightness is often sacrificed to improve efficiency (e.g., sawtooth approximation). In this paper, we describe a new approach to efficiently compute tighter bounds by i) conducting a prioritized breadth first search over the reachable beliefs, ii) propagating upper bound improvements with an augmented POMDP and iii) using exact linear programming (instead of the sawtooth approximation) for upper bound interpolation. As a result, we can represent the bounds more compactly and significantly reduce the gap between upper and lower bounds on several benchmark problems. |
| 关 键 词: | 最优值函数; 优先广度; 线性规划 |
| 课程来源: | 视频讲座网 |
| 最后编审: | 2020-05-31：吴雨秋(课程编辑志愿者) |
| 阅读次数: | 68 |