BriefGPT.xyz
May, 2019
回归任务中的不确定性预测评估和校准
Evaluating and Calibrating Uncertainty Prediction in Regression Tasks
HTML
PDF
Dan Levi, Liran Gispan, Niv Giladi, Ethan Fetaya
TL;DR
本文提出了一种新的针对回归任务中不确定性预测校准的方法和评估方法,并通过对合成问题和对COCO和KITTI数据集的物体检测边界框回归任务的实验验证,展示出基于直方图的聚类方法和基于缩放的校准方法的效果相当好。
Abstract
Predicting not only the target but also an accurate measure of uncertainty is important for many applications and in particular safety-critical ones. In this work we study the
calibration
of
uncertainty prediction
→