An Additive Reinforcement Learning-猫眼课题宝

DOI：

10.1007/978-3-642-04274-4_63

发表时间：

2009-09

期刊：

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Takeshi Mori;S. Ishii

在强化学习中，准备基函数需要一定的先验知识，通常是一项艰巨的任务。为了克服这个困难，Keller 等人提出了一种自适应基函数构造技术。最近，但它需要过多的计算成本。我们针对这种情况提出了一种有效的方法，其中近似值函数的问题自然地分解为许多子问题，每个子问题都可以以较小的计算成本来解决。计算机实验表明，我们的方法所需的CPU时间比现有方法小得多。

In reinforcement learning, preparing basis functions requires a certain amount of prior knowledge and is in general a difficult task. To overcome this difficulty, an adaptive basis function construction technique has been proposed by Keller et al. recently, but it requires excessive computational cost. We propose an efficient approach to this context, in which the problem of approximating the value function is naturally decomposed into a number of sub-problems, each of which can be solved at small computational cost. Computer experiments show that the cpu-time needed by our method is much smaller than that of the existing method.

An Additive Reinforcement Learning

文献摘要

相似文献