Q$5$: Please explain why the following algorithmic implementation of the convolution layer is not

considered hardware$-$efficient.

for $($n$=0$; n max$)\{$

max $=$ I$[$n$][$m$][$Ux$+$i$][$Uy$+$j$]$;

$\}$

$\}$

$\}$

O$[$n$][$m$][$x$][$y$]=$ max;

$\}$

$\}$

$\}$

$\}$