[kwrobel]

"As Jeremy Howard points out, even academic papers often use softmax for multi-class classification, and I too have already seen it used incorrectly in blogs and papers during my short time studying DL."

AFAIK softmax should be used with mutli-class classification and sigmoid can be used with mutli-label classification.

[qiemem]

I think the author meant multi-label.

[king_magic]

Exactly. I’m pretty confused about the excerpt, it struck me as strange as soon as I read it - maybe it’s a typo?

[nafizh]

I think the author meant sigmoid.