In this paper, we focus on estimating the probabilistic upper bounds of star discrepancy for Hilbert space filling curve (HSFC) sampling. The main idea is the stratified random sampling method, but the strict condition for sampling number $N=m^d$ of jittered sampling is removed. We inherit the advantages of this sampling and get better results than Monte Carlo (MC) sampling.
翻译:在本文中,我们侧重于估算Hilbert空间填充曲线(HSFC)取样中恒星差异的概率上限。主要想法是分层随机取样方法,但排除了采样数字($N=m ⁇ d$)的严格条件。我们继承了这种采样的优势,并取得了比蒙特卡洛(Monte Carlo)取样更好的结果。