arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.11961cs.LGstat.ML
收藏

锐利校准的高斯过程

Sharp Calibrated Gaussian Processes

Technical University of Munich(慕尼黑工业大学);University of British Columbia(不列颠哥伦比亚大学);Vector Institute(向量研究院)

Alexandre Capone, Geoff Pleiss, Sandra Hirche

AI总结:

针对高斯过程不确定性校准不佳的问题,提出一种基于不同超参数满足经验校准约束的灵活校准方法,生成更紧凑的预测分位数,在合理假设下实现校准并在锐利性上优于现有方法。

摘要:

虽然高斯过程是各种工程和科学应用的主要工具,但其不确定性估计不满足频率学保证,在实践中可能校准不当。最先进的校准模型设计方法依赖于膨胀高斯过程后验方差,这产生的置信区间可能过于粗糙。为解决此问题,我们提出一种校准方法,该方法使用受普通高斯过程后验方差启发的计算来生成预测分位数,但使用一组不同的超参数,以满足经验校准约束。这产生了一种比现有方法更灵活的校准方法,我们对其进行优化以生成紧凑的预测分位数。我们的方法在合理假设下被证明能产生校准模型。此外,在校准回归中,它在锐利性方面优于现有方法。

英文摘要:

While Gaussian processes are a mainstay for various engineering and scientific applications, the uncertainty estimates don't satisfy frequentist guarantees and can be miscalibrated in practice. State-of-the-art approaches for designing calibrated models rely on inflating the Gaussian process posterior variance, which yields confidence intervals that are potentially too coarse. To remedy this, we present a calibration approach that generates predictive quantiles using a computation inspired by the vanilla Gaussian process posterior variance but using a different set of hyperparameters chosen to satisfy an empirical calibration constraint. This results in a calibration approach that is considerably more flexible than existing approaches, which we optimize to yield tight predictive quantiles. Our approach is shown to yield a calibrated model under reasonable assumptions. Furthermore, it outperforms existing approaches in sharpness when employed for calibrated regression.

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。

↑