一句话：提出操作校准，将 DNN 分类器置信度函数针对运行域校准，以减少现场高置信误判导致的损失。
训练好的 DNN 模型日益成为软件系统的一部分，但在现场常因运行数据与训练数据存在不可避免的微小偏差而表现不佳。尤其危险的是模型会以高置信度给出错误预测，造成严重损失。为此，论文提出“操作校准”（operational calibration）：面向 DNN 分类器的运行域校准其置信度函数，以最小化不准确置信度带来的损失，并将其视为整个系统工程中的必要调试步骤。摘要主要阐述问题与概念定位，未给出具体校准算法、实验设置或量化结果。
Trained DNN models are increasingly adopted as integral parts of software systems, but they often perform deficiently in the field. A particularly damaging problem is that DNN models often give false predictions with high confidence, due to the unavoidable slight divergences between operation data and training data. To minimize the loss caused by inaccurate confidence, operational calibration, i.e., calibrating the confidence function of a DNN classifier against its operation domain, becomes a necessary debugging step in the engineering of the whole system.
@inproceedings{paperbot5753,
title = {Operational calibration: debugging confidence errors for DNNs in the field},
author = {Zenan Li and Xiaoxing Ma and Chang Xu and Jingwei Xu and Chun Cao and Jian Lü},
booktitle = {FSE},
year = {2020},
doi = {10.1145/3368089.3409696}
}