arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1910.09328cs.LGstat.ML
收藏

线性区域约束下的高斯积分

Integrals over Gaussians under Linear Domain Constraints

University of Tuebingen(图宾根大学);MPI for Intelligent Systems(马克斯·普朗克智能系统研究所);Technical University of Munich(慕尼黑工业大学)

Alexandra Gessner, Oindrila Kanjilal, Philipp Hennig

摘要:

线性约束多变量高斯密度的积分是机器学习和统计学中的常见问题,出现在广义线性模型和贝叶斯优化等任务中。然而,它们是出了名的难以计算,并且使问题进一步复杂化的是,这样的积分的数值可能非常小。我们提出了一个有效的黑箱算法,利用几何估计的积分在一个小的,截断高斯体积,并从中模拟。我们的算法使用Holmes-Diaconis-Ross(HDR)方法结合椭圆切片采样(ESS)的分析版本。适应线性设置,ESS允许无拒绝采样,因为椭圆和域边界的交叉点有封闭形式的解决方案。HDR的核心思想是通过使用一系列嵌套域将积分分解为计算器计算的条件概率。值得注意的是,它允许直接计算积分值的对数,从而能够计算极小的概率质量。我们证明了HDR和ESS的定制组合在高维积分和贝叶斯优化的熵搜索上的有效性。

英文摘要:

Integrals of linearly constrained multivariate Gaussian densities are a frequent problem in machine learning and statistics, arising in tasks like generalized linear models and Bayesian optimization. Yet they are notoriously hard to compute, and to further complicate matters, the numerical values of such integrals may be very small. We present an efficient black-box algorithm that exploits geometry for the estimation of integrals over a small, truncated Gaussian volume, and to simulate therefrom. Our algorithm uses the Holmes-Diaconis-Ross (HDR) method combined with an analytic version of elliptical slice sampling (ESS). Adapted to the linear setting, ESS allows for rejection-free sampling, because intersections of ellipses and domain boundaries have closed-form solutions. The key idea of HDR is to decompose the integral into easier-to-compute conditional probabilities by using a sequence of nested domains. Remarkably, it allows for direct computation of the logarithm of the integral value and thus enables the computation of extremely small probability masses. We demonstrate the effectiveness of our tailored combination of HDR and ESS on high-dimensional integrals and on entropy search for Bayesian optimization.

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。

↑