Re:T0* – Series of encoder-decoder models trained on a large...
by krastanov on Monday October 18, 2021 @11:14AM
(#28906246)
This is kinda missing the point here... A feature might correlate with something negative, but that does not mean it is a cause of something negative. Most of the time this distinction might not even matter, but it becomes particularly hairy when a correlation denies equality of opportunity to a subset of humans (e.g., in the hiring example of a sibling comment),
[ Parent
| on HN ]
There are no comments on this comment yet.