Mostafa ElAraby, Sabyasachi Sahoo, Yann Pequignot, Paul Novello, Liam Paull
TL;DR使用渐变空间和类原型增强深度神经网络(DNN)在图像分类中的离群样本检测
Abstract
deep neural networks (DNNs) often fail silently with over-confident
predictions on out-of-distribution (OOD) samples, posing risks in real-world
deployments. Existing techniques predominantly emphasize either the feature
representation space or the gradient norms computed with respect