this feels like at best a pretty misleading and insensitive comment; 140 characters is limiting, but being more precise might be helpful
-
Every ML alg I've worked with over-predicts the majority class. In many classes, that leads to discrimination against human minorities
That's poor calibration, not bias. (To solve a problem we need to first diagnose it correctly.)
-
-
So you claim no ML algorithm has discriminated because EVERYONE properly calibrates their ML algorithms?
-
No, most people don't but that does not mean that the algorithm discriminates, but that the designers are incapable.
- Näytä vastaukset
Uusi keskustelu -
Lataaminen näyttää kestävän hetken.
Twitter saattaa olla ruuhkautunut tai ongelma on muuten hetkellinen. Yritä uudelleen tai käy Twitterin tilasivulla saadaksesi lisätietoja.