1
$\begingroup$

Say we have a relationship $ z = Wx$ for a multi layer perceptron where $z$ and $x$ are $n$ dimensional vectors. When we find $\frac{dz}{dx}$ , I would assume this would just be $W$, not $W^T$.

I was wondering where this result comes from - I can't figure out why this makes sense.

Improve this question
$\endgroup$
1
  • 2
    $\begingroup$ It's a convention. You can use either convention (numerator convention or denominator convention) as long as you're consistent. $\endgroup$
    – Arya McCarthy
    Commented Apr 10, 2022 at 23:01

0

Reset to default