图 书 馆
线性可解MDPs理论的推广Generalizations of the Theory of Linearly Solvable MDPs |
|
| 课程网址: | http://videolectures.net/cyberstat2012_dvijotham_linearly_solvabl... |
| 主讲教师: | Krishnamurthy Dvijotham |
| 开课单位: | 华盛顿大学 |
| 开课时间: | 2012-10-16 |
| 课程语种: | 英语 |
| 中文简介: |  |
| 课程简介: | We present various generalizations of the theory of linearly solvable optimal control. One generalization is the extension to the game theoretic and risk-sensitive setting, which requires replacing KL costs with Renyi divergences. Another extension is the derivation of efficient policy gradient algorithms that only need one to sample the state space (rather than the state-action space) for various problem formulations (finite horizon, infinite horizon, first-exit, discounted). For the finite horizon case, we show that the PI^2 algorithm (Policy Improvement with Path Integrals) can be seen as a special case of our policy gradient algorithms when applied to a risk-seeking objective. Finally, we present applications of these policy gradient algorithms to various problems in movement control and power systems. |
| 关 键 词: | 线性可解; 最优控制理论; 策略梯度算法; 游戏理论; 风险敏感设置 |
| 课程来源: | 视频讲座网 |
| 最后编审: | 2020-10-26：zyk |
| 阅读次数: | 91 |