目录
谷歌大脑提出概念激活向量,助力神经网络可解释性研究
https://github.com/tensorflow/tcav
Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors