Polynomial time approximate or perfect samplers for discretized Dirichlet distribution
Polynomial time approximate or perfect samplers for discretized Dirichlet distribution
复制标题
离散狄利克雷分布的多项式时间近似或完美采样器
DOI：
10.1007/s13160-010-0002-0
复制
发表时间：
2010
影响因子：
0.9
通讯作者：
and S. Kijima
中科院分区：
文献类型：
--
作者：
T. Matsui;M.Motoki;N.Kamatani;and S. Kijima
在本文中，我们提出了两个马尔可夫链的抽样随机向量分布根据离散Dirichlet分布。它们的混合速率分别以O（n2 log Δ）和O（n3 log Δ）为界，其中n为维数，1/Δ为离散化的网格尺寸。我们的第二个链给出了一个完美的采样器，它是基于过去的单调耦合。我们还报告了模拟的结果。
In this paper, we propose two Markov chains for sampling random vectors that are distributed according to discretized Dirichlet distribution. Their mixing rates are bounded by O(n2logΔ) and O(n3logΔ), wherenis the dimension and 1/Δis the grid size for discretization. Our second chain gives a perfect sampler that is based on monotone coupling from the past. We also report the results of simulations.