多类别分类器的置信度校准

Nov, 2024

Confidence Calibration of Classifiers with Many Classes

Adrien LeCoz, Stéphane Herbin, Faouzi Adjed

TL;DR本研究针对基于神经网络的分类模型，提出了一种新的方法来解决多类别分类器置信度校准的问题。通过将多类别校准转化为单个替代二分类器的校准，该方法提升了传统校准方法的效率。实验表明，该方法显著改善了现有校准技术在图像和文本分类任务中的表现。

Abstract

For classification models based on Neural Networks, the maximum predicted class probability is often used as a confidence score. This score rarely predicts well the probability of making a correct prediction and requires a →