arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2010.01760stat.CO
收藏

无限切片采样

Unbounded Slice Sampling

Daichi Mochihashi

摘要:

切片抽样是一种有效的马尔可夫链蒙特卡罗算法,它可以从非归一化密度中抽样,并且接受率总是$1$。然而,当要采样的变量是无界的时,它的“逐步退出”启发式仅在局部起作用,使得难以均匀地探索可能的候选者。本文提出了一种简单的变元法,从[0,1)等价地对无界变量进行切片采样。

英文摘要:

Slice sampling is an efficient Markov Chain Monte Carlo algorithm to sample from an unnormalized density with acceptance ratio always $1$. However, when the variable to sample is unbounded, its "stepping-out" heuristic works only locally, making it difficult to uniformly explore possible candidates. This paper proposes a simple change-of-variable method to slice sample an unbounded variable equivalently from [0,1).

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。

↑