BriefGPT.xyz
Aug, 2019
基于 bin 的温度缩放（BTS）：通过简单的缩放技术提高置信度校准性能
Bin-wise Temperature Scaling (BTS): Improvement in Confidence Calibration Performance through Simple Scaling Techniques
HTML
PDF
Byeongmoon Ji, Hyemin Jung, Jihyeun Yoon, Kyungyul Kim, Younghak Shin
TL;DR
本研究采用温度缩放方法构建精细的分组缩放以及采用验证样本增强方法，取得了各种深度卷积神经网络模型在多个数据集上的一致且优异的校准表现。
Abstract
The
prediction reliability
of
neural networks
is important in many applications. Specifically, in safety-critical domains, such as cancer prediction or autonomous driving, a reliable confidence of model's predict
→