BriefGPT.xyz
Dec, 2020
针对领域偏移场景的后验不确定性校准
Post-hoc Uncertainty Calibration for Domain Drift Scenarios
HTML
PDF
Christian Tomani, Sebastian Gruber, Muhammed Ebrar Erdem, Daniel Cremers, Florian Buettner
TL;DR
本文探讨了深度神经网络中不确定性校准问题，并提出了一种针对域偏移的后处理校准方法，其通过对验证集的样本进行扰动，可大幅提高模型的校准性能。
Abstract
We address the problem of
uncertainty calibration
. While standard deep
neural networks
typically yield uncalibrated predictions, calibrated confidence scores that are representative of the true likelihood of a pr
→