提出新型核方法，快速准确检验概率模型校准性。
Fast and Scalable Score-Based Kernel Calibration Tests
- 基于得分构造新核函数，无需密度样本即可估计
- 控制第一类错误，避免昂贵期望近似
- 适合需要高效校准验证的机器学习应用
我们提出了基于核的校准条件斯坦离散度检验（KCCSD检验），一种用于评估具有明确定义得分的概率模型校准性的非参数、基于核的方法。与以往方法不同，该检验避免了可能代价高昂的期望近似，同时保证了第一类错误的控制。通过使用一类新的基于得分的概率核函数，可无需概率密度样本进行估计，并采用条件拟合优度准则来构建KCCSD检验的U统计量，实现了上述改进。我们在多种合成场景下验证了该检验的性质。
原文摘要 · Abstract (English)
We introduce the Kernel Calibration Conditional Stein Discrepancy test (KCCSD test), a non-parametric, kernel-based test for assessing the calibration of probabilistic models with well-defined scores. In contrast to previous methods, our test avoids the need for possibly expensive expectation approximations while providing control over its type-I error. We achieve these improvements by using a new family of kernels for score-based probabilities that can be estimated without probability density samples, and by using a conditional goodness-of-fit criterion for the KCCSD test's U-statistic. We demonstrate the properties of our test on various synthetic settings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。