2 ms·
I do agree with you that it is fine, what I was getting at was that there needs to be a way to measure uncertainty in a manner that is robust to unbalanced dist
by PartiallyTyped 4y ago
I do agree with you that it is fine, what I was getting at was that there needs to be a way to measure uncertainty in a manner that is robust to unbalanced distributions or context drifting.
- elcomet 4y agoOkay I see what you mean, I agree with you on this.