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~~

面向预测的高斯过程后验

Predictively Oriented Gaussian Process Posteriors

Callum Lau, Jeremias Knoblauch, Louis Sharrock

arXiv 2610.03201首次发表:更新:

发表机构

University College London(伦敦大学学院)

机构由 AI 辅助整理,请以论文原文为准。

AI 中文总结

本文提出面向预测的高斯过程(PrO-GPs),以预测不确定性为推断目标,通过简化公式和采样方案高效计算后验,实验证明在模型错误指定下其预测分布校准优于标准高斯过程。

AI 中文摘要

高斯过程(GPs)是建模和量化函数关系中不确定性的强大工具。然而,它们要求实践者做出许多设计决策,例如核函数和观测模型的选择。次优选择可能产生未正确指定的模型,这些模型无法捕捉底层数据生成过程。我们引入了面向预测的高斯过程(PrO-GPs),它将预测不确定性视为主要推断目标,并提供了标准高斯过程的稳健替代方案。尽管非参数模型的PrO后验直接计算是不可行的,但我们推导出了一个简化公式和实用的采样方案,以实现高效计算。通过合成和真实数据实验,我们表明,在模型错误指定下,与标准高斯过程方法相比,PrO-GPs能产生校准更好的预测分布。

英文摘要

Gaussian Processes (GPs) are a powerful tool for modelling and quantifying uncertainty in functional relationships. However, they require practitioners to make a number of design decisions, such as the choice of the kernel and the observation model. Suboptimal choices can produce misspecified models that do not capture the underlying data generating process. We introduce Predictively Oriented Gaussian Processes (PrO-GPs), which treat predictive uncertainty as the primary inferential target and provide a robust alternative to standard GPs. Although direct computation of a PrO posterior for nonparametric models is intractable, we derive a reduced formulation and practical sampling scheme for efficient computation. Through synthetic and real data experiments, we show that PrO-GPs produce better calibrated predictive distributions under model misspecification compared to standard GP approaches.

论文原文

arXiv 摘要页 · PDF 原文 · HTML 原文

↑