Re:.98 AUC classifying race from chest x-rays (external vali...
by nightcracker on Wednesday August 04, 2021 @04:08PM
(#28065660)
The point is that if a neural network can predict the relationship when labelled with such accuracy it means this information is unavoidably encoded within the image. The result is that even when race is not labeled that the AI might still be racist by automatically (and unknowingly) learning the race.
[ Parent
| on HN ]
Re:.98 AUC classifying race from chest x-rays (external vali...
by pvitz on Tuesday August 10, 2021 @05:40AM
(#28126817)
If one only trains the model to correctly identify a disease (e.g. maximising the correct identification and minimising the error), then there is no space for the model to be "racist". I don't question that there is a signal inside the images, but someone has to change the model and training to then maliciously use that information.