图 书 馆
考虑到看不见的状态因素的强化学习是不可能的Considering Unseen States as Impossible in Factored Reinforcement Learning |
|
| 课程网址: | http://videolectures.net/ecmlpkdd09_kozlova_cusifr/ |
| 主讲教师: | Olga Kozlova |
| 开课单位: | 皮埃尔与玛丽居里大学 |
| 开课时间: | 2009-10-20 |
| 课程语种: | 英语 |
| 中文简介: |  |
| 课程简介: | The Factored Markov Decision Process (FMDP) framework is a standard representation for sequential decision problems under uncertainty where the state is represented as a collection of random variables. Factored Reinforcement Learning (FRL) is an Model-based Reinforcement Learning approach to FMDPs where the transition and reward unctions of the problem are learned. In this paper, we show how to model in a theoretically well-founded way the problems where some combinations of state variable values may not occur, giving rise to impossible states. Furthermore, we propose a new heuristics that considers as impossible the states that have not been seen so far. We derive an algorithm whose improvement in performance with respect to the standard approach is illustrated through benchmark experiments. |
| 关 键 词: | 计算机科学; 马尔可夫决策; 强化学习 |
| 课程来源: | 视频讲座网 |
| 最后编审: | 2019-12-04：lxf |
| 阅读次数: | 63 |