BriefGPT.xyz
Oct, 2019
多类别分类中的校准测试：一个统一的框架
Calibration tests in multi-class classification: A unifying framework
HTML
PDF
David Widmann, Fredrik Lindsten, Dave Zachariah
TL;DR
文章讨论了多类别分类中的概率模型拟合和校准问题，并提出了一种基于矩阵核的估计器来解释校准度量的测试统计量。
Abstract
In safety-critical applications a
probabilistic model
is usually required to be calibrated, i.e., to capture the uncertainty of its predictions accurately. In
multi-class classification
,
→