Hey there, I don’t understand why this the answer is false there
{Moderator’s Edit: Quiz Solution Removed}
The way I see it is that b is canceled during the Z normalize computation since it’s a constant factor the mean’s substraction takes it away, so it’s false that it’s substracted when computing Z̃ tilde
What am I missing there ?
Hey @MadeInShineA,
Welcome, and we are glad that you could become a part of our community 
When we compute \tilde{z}^{[l]}, the values are re-centred at the origin, as you have pointed out. Now, let’s say that the mean without b for the output values is 6, and now with b, it is 7, but since batch-normalization will reduce the mean to 0 in both the cases, doesn’t that mean in the second case, mean = 6 (in the first case) and b is subtracted to reduce the mean to 0?
The thing to note here is that the question doesn’t ask if b is explicitly subtracted in the formula for batch-normalization, it instead asks whether the output will have any difference if we drop the bias parameter from the forward propagation equations. I hope this resolves your query.
P.S. - It is against the community guidelines to share the solutions publicly. Please keep this in mind for future references. No harm done right now, I have edited your post to remove the solution contained within it.
Cheers,
Elemento
Oh sorry… I didn’t know it wasn’t allowed, I will keep that in mind for the next time!
Thanks for your quick response!
So that’s what I thought, I went to far thinking the quiz was trying to trick me 
Cheers !