Loss function for dnorm in predcls about sgg HOT 3 CLOSED

I got it. The change for $(FG+BG)/FG$ will enlarge the loss of large graph.
Sorry to bother you.

from sgg.

Hi, I'm glad you've resoved this issue. You are right in your last comment, but I just wanted to add that we tested the variant when the edge loss weight is fixed for all batches (equivalent to adjusting the learning rate as you mentioned). Such a variant also worked well in some cases (see the paper for details). The weight of our edge loss adapts to each batch of images/scene graphs and does not require tuning for a new dataset (where scene graphs may have different sparsity). So these properties of our loss can be helpful in practice.

from sgg.

It really addressed my concerns. Thank you for your kind reply.

from sgg.