Week 3
Machine Learning Motivation
In the video, the first hidden layer contains 3 neurons, but the outputs are labeled a_1, a_2, …, a_n, using the same ‘n’ as for the input features x_1, …, x_n. Is this intentional, or could it be confusing, since the number of inputs and neurons is not necessarily the same? Should it perhaps be a_3 instead of a_n here?
As a reminder, the way that worked was that you have
a dataset made up of some set
of features you called x_1,
x_2, all the way up to x_n
Just to be clear, these a superscript ones are now matrices containing all the a vectors from the previous layer, this a_1, a_2, and so on to a_n.