loss-function
an archive of posts with this tag
-
Why cross entropy, of all things: from BCE to KL divergence
Why classification uses cross entropy instead of MSE, traced as a single line from information content and entropy through BCE and MLE to KL divergence.