Hacker News new | ask | show | jobs
What Explainable AI fails to explain (and how we fix that) (towardsdatascience.com)
3 points by alvinwan 2257 days ago
1 comments

tl;dr We made models as accurate as neural networks and as interpretable as decision trees. This work focuses on image classification for computer vision. You can also find out more on our project page http://nbdt.alvinwan.com or arxiv submission https://arxiv.org/abs/2004.00221