BriefGPT.xyz
Aug, 2022
卷积权重的奇异值分解:一种CNN可解释性框架
The SVD of Convolutional Weights: A CNN Interpretability Framework
HTML
PDF
Brenda Praggastis, Davis Brown, Carlos Ortiz Marrero, Emilie Purvine, Madelyn Shapiro...
TL;DR
本文提出一种基于张量的奇异值分解的方法,用于理解卷积层的动态过程和发现卷积特征之间的相关性,以及在图像分类网络中应用超图模型进行可解释性研究。
Abstract
deep neural networks
used for image classification often use
convolutional filters
to extract distinguishing features before passing them to a linear classifier. Most
→