Week 3 Ass. 2: Dropout in Conv. Nets?

The technique of Dropout was discussed in the context of fully connected layers. However, it appears all of a sudden in the assignment on U-Net in the convolutional context. Could you please add some discussion of Dropout in this context?

Hi Meir,

The original paper states the following:

“Drop-out layers at the end of the contracting path perform further implicit data augmentation”.

This is subtle and points to a different function of drop-out layers than regularization. Maybe this is why it’s left out of the discussion at this point.

1 Like