laura-rieger/deep-explanation-penalization

Code for using CDEP from the paper "Interpretations are useful: penalizing explanations to align neural networks with prior knowledge" https://arxiv.org/abs/1909.13584

语言：Jupyter Notebook · GitHub 星标：126 · Fork：15

话题：ai · artificial-intelligence · cdep · convolutional-neural-network · data-science · deep-learning · explainability · explainable-ai

在 GitHub 上查看原仓库