3 ms·
The classic (multi-decade problem repeated again, and again, and again...) is face detection. Face detection models continue to get trained by primarily white d
by trobertson 6y ago
The classic (multi-decade problem repeated again, and again, and again...) is face detection. Face detection models continue to get trained by primarily white developers/researches who continue to use primarily white people as their training set. And then they're shocked when their model performs terribly for everyone who isn't of European decent.
This case is so pervasive that it gets taught in basically every university that teaches ML - and yet those students go on to repeat the exact problem they were educated against.
- visarga 6y ago> Face detection models continue to get trained by primarily white developers/researches who continue to use primarily white people as their training set. I guess you didn't watch Asian developments in face recognition closely in the last few years.
- aparsons 6y agoLet’s pause all ML research until we have 50% black representation!
- tarsinge 6y agoA lighthearted example: https://www.kotaku.co.uk/2017/01/12/how-we-accidentally-made-a-racist-videogame https://www.kotaku.co.uk/2017/01/12/how-we-accidentally-made...
- spoopyskelly 6y agoDoes the face recognition tech in China have big problems with european faces?