arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.00393cs.LGstat.ML
收藏

域间观测值的分层诱导点高斯过程

Hierarchical Inducing Point Gaussian Process for Inter-domain Observations

Columbia University(哥伦比亚大学);The Observatories of the Carnegie Institution for Science(卡内基科学研究所天文台)

Luhuan Wu, Andrew Miller, Lauren Anderson, Geoff Pleiss, David Blei, John Cunningham

摘要:

我们研究了域间高斯过程(GP)的一般问题:GP实现和该实现的噪声观测位于不同域的问题。当这些域之间的映射是线性的,如积分或微分,推理仍然是封闭形式。然而,我们社区开发的许多缩放和近似技术并不适用于此设置。在这项工作中,我们引入了层次诱导点GP(HIP-GP),一个可扩展的域间GP推理方法,使我们能够提高近似精度,通过增加诱导点的数量以百万计。HIP-GP依赖于具有网格结构的诱导点和稳定的核假设,适用于低维问题。在开发HIP-GP中,我们引入了(1)一个快速白化策略,以及(2)一个新的共轭梯度预处理器,它可以在一般的GP设置中有所帮助。我们的代码可以在https://github.com/cunningham-lab/hipgp上找到。

英文摘要:

We examine the general problem of inter-domain Gaussian Processes (GPs): problems where the GP realization and the noisy observations of that realization lie on different domains. When the mapping between those domains is linear, such as integration or differentiation, inference is still closed form. However, many of the scaling and approximation techniques that our community has developed do not apply to this setting. In this work, we introduce the hierarchical inducing point GP (HIP-GP), a scalable inter-domain GP inference method that enables us to improve the approximation accuracy by increasing the number of inducing points to the millions. HIP-GP, which relies on inducing points with grid structure and a stationary kernel assumption, is suitable for low-dimensional problems. In developing HIP-GP, we introduce (1) a fast whitening strategy, and (2) a novel preconditioner for conjugate gradients which can be helpful in general GP settings. Our code is available at https: //github.com/cunningham-lab/hipgp.

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。

↑