Polynomial Time Approximate Sampler for Discretized Dirichlet Distribution

Polynomial Time Approximate Sampler for Discretized Dirichlet Distribution
复制标题

离散狄利克雷分布的多项式时间近似采样器

DOI:
10.1007/978-3-540-24587-2_69
复制
发表时间:
2003
期刊:
影响因子:
3.9
通讯作者:
N. Kamatani
N. Kamatani
中科院分区:
医学2区
文献类型:
--
作者:
Tomomi Matsui;Mitsuo Motoki;N. Kamatani

文献摘要

参考文献

被引文献

相似文献

在本文中,我们提出了一种马尔可夫链,用于对根据离散狄利克雷分布分布的随机向量进行采样。我们证明我们的马尔可夫链正在快速混合,也就是说,我们的链的混合时间以 (1/2)n(n - 1)ln((δ - n)e − 1) 为界,其中 n 是维度(参数数量),1/Δ 是离散化的网格大小,e 是误差界限。因此,获得的界限不依赖于参数的大小,并且与网格大小的倒数的对数成线性。我们使用路径耦合方法来估计混合时间。我们还通过实验展示了我们的链的收敛速度。
In this paper, we propose a Markov chain for sampling a random vector distributed according to a discretized Dirichlet distribution. We show that our Markov chain is rapidly mixing, that is, the mixing time of our chain is bounded by (1/2)n(n - 1)ln((δ - n)e − 1) where n is the dimension (the number of parameters), 1/Δ is the grid size for discretization, and e is the error bound. Thus the obtained bound does not depend on the magnitudes of parameters and linear to the logarithm of the inverse of grid size. We estimate the mixing time by using the path coupling method. We also show the rate of convergence of our chain experimentally.
DOI: 10.1086/338446
发表时间: 2002-01-01
影响因子: 9.8
作者:
Niu, TH;Qin, ZHS;Liu, JS
通讯作者: Liu, JS