arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1206.3291cs.AI
收藏

分层POMDP控制器的似然最大化优化

Hierarchical POMDP Controller Optimization by Likelihood Maximization

TU Berlin(柏林工业大学);University of Toronto(多伦多大学);University of Waterloo(滑铁卢大学)

Marc Toussaint, Laurent Charlin, Pascal Poupart

摘要:

计划通常可以通过将任务分解为按层次安排的更小的任务来简化。Charlin等人。[4]最近表明,层次发现问题可以被框定为非凸优化问题。然而,解决这样的优化问题的固有计算困难使得难以扩展到现实世界的问题。在另一条研究路线中,Toussaint等人[18]开发了一种通过最大似然估计来解决规划问题的方法。在本文中,我们将展示如何层次发现问题,部分可观察域可以使用类似的最大似然方法来解决。我们的技术rst转换成一个动态的贝叶斯网络,通过它可以自然地发现一个层次结构,同时优化政策的问题。实验结果表明,该方法的规模优于以前的技术的基础上非凸优化。

英文摘要:

Planning can often be simpli ed by decomposing the task into smaller tasks arranged hierarchically. Charlin et al. [4] recently showed that the hierarchy discovery problem can be framed as a non-convex optimization problem. However, the inherent computational di culty of solving such an optimization problem makes it hard to scale to realworld problems. In another line of research, Toussaint et al. [18] developed a method to solve planning problems by maximumlikelihood estimation. In this paper, we show how the hierarchy discovery problem in partially observable domains can be tackled using a similar maximum likelihood approach. Our technique rst transforms the problem into a dynamic Bayesian network through which a hierarchical structure can naturally be discovered while optimizing the policy. Experimental results demonstrate that this approach scales better than previous techniques based on non-convex optimization.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑