https://pair.withgoogle.com/explorables/saliency/
Searching for Unintended Biases With Saliency
Astrid Bertrand, Adam Pearce and Nithum Thain
December 2022
“If there are spurious correlations in the training data, the model might develop unintended biases that could lead to mistakes. In this post, we explore a technique for detecting these biases by asking for an explanation of how models make decisions.”