Att känna in sig i en aktiveringsfunktion

Jeremy Howard på fast.ai uppmanar till att man ska känna in sig de matematiska funktioner som man använder för deep learning.

In this case, we'll want to be able to predict multiple things.

And so Softmax wouldn't be good, because Softmax doesn't *like* predicting multiple things

And like, I would definitely recommend anthropomorphizing your activation functions, right.

They have personalities, okey, and the personality of the Softmax is it *wants to pick a thing* and people forget this all the time.

Jämför med inlägget Empati med tingen

/Simon

infontology, maskiner, mentala_modeller, empati