The effect of the choice of neural network depth and breadth on the size of its hypothesis space

06/06/2018

∙

by   Lech Szymanski, et al.

∙

0

∙

share

We show that the number of unique function mappings in a neural network hypothesis space is inversely proportional to ∏_lU_l!, where U_l is the number of neurons in the hidden layer l.