arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2307.03093cs.LGstat.ML
收藏

超越直觉:将 Gaussian Processes(GPs,高斯过程)应用于真实世界数据的框架

Beyond Intuition, a Framework for Applying GPs to Real-World Data

University of Cambridge(剑桥大学);British Antarctic Survey(英国南极调查局);Max Planck Institute for Intelligent Systems(马克斯·普朗克智能系统研究所);California Institute of Technology(加州理工学院);Aalto University(阿尔托大学)

Kenza Tazi, Jihao Andreas Lin, Ross Viljoen, Alex Gardner, ST John, Hong Ge, Richard E. Turner

AI总结:

本文提出一个判断 GP 适用性并规范建模流程的框架,强调核设计与可扩展性,且在冰川高程变化案例中提升了测试精度。

摘要:

Gaussian Processes(GPs,高斯过程)为在小型、结构化且相关的数据集上进行回归提供了一种有吸引力的方法。然而,其部署受到计算成本以及在简单低维数据集之外如何应用 GPs 的指导有限的阻碍。我们提出一个框架,用于识别 GPs 对给定问题的适用性,以及如何建立稳健且规范良好的 GP 模型。这些指南将经验丰富的 GP 从业者的决策形式化,重点强调核设计和计算可扩展性选项。随后,该框架被应用于冰川高程变化的案例研究,并在测试时获得了更准确的结果。

英文摘要:

Gaussian Processes (GPs) offer an attractive method for regression over small, structured and correlated datasets. However, their deployment is hindered by computational costs and limited guidelines on how to apply GPs beyond simple low-dimensional datasets. We propose a framework to identify the suitability of GPs to a given problem and how to set up a robust and well-specified GP model. The guidelines formalise the decisions of experienced GP practitioners, with an emphasis on kernel design and options for computational scalability. The framework is then applied to a case study of glacier elevation change yielding more accurate results at test time.

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。

↑