arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1601.08057stat.COstat.MEstat.ML
收藏

关于哈密顿蒙特卡洛的几何遍历性

On the Geometric Ergodicity of Hamiltonian Monte Carlo

Department of Statistical Science, University College(大学学院统计科学系);Department of Statistics, University of Warwick(华威大学统计系);The Alan Turing Institute(艾伦图灵研究所)

Samuel Livingstone, Michael Betancourt, Simon Byrne, Mark Girolami

AI总结:

本文建立了哈密顿蒙特卡洛方法几何遍历性的充分与必要条件,并探讨了积分时间选择对尾部行为的影响,为实践提供指导。

摘要:

我们建立了哈密顿蒙特卡洛方法产生的马尔可夫链在何种一般条件下会或不会具有几何遍历性。我们考虑了位置无关和位置相关的积分时间两种实现方式。在前一种情况下,我们发现几何遍历性的条件本质上是对数密度的梯度,该梯度渐近地指向空间中心且增长不超过线性速度。在积分时间允许在空间不同区域变化的理想化场景中,我们表明对于更广泛的尾部行为类别,可以恢复几何遍历性,这为实践中该自由参数的选择提供了一些指导原则。

英文摘要:

We establish general conditions under which Markov chains produced by the Hamiltonian Monte Carlo method will and will not be geometrically ergodic. We consider implementations with both position-independent and position-dependent integration times. In the former case we find that the conditions for geometric ergodicity are essentially a gradient of the log-density which asymptotically points towards the centre of the space and grows no faster than linearly. In an idealised scenario in which the integration time is allowed to change in different regions of the space, we show that geometric ergodicity can be recovered for a much broader class of tail behaviours, leading to some guidelines for the choice of this free parameter in practice.

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。

↑