#StackBounty: #neural-networks #deep-learning #normalization Purpose of L2 normalization for triplet network

Bounty: 100

Triplet-based distance learning for face recognition (http://arxiv.org/abs/1503.03832) seems very effective. I’m curious about one particular aspect of the paper. As part of finding an embedding for a face, the authors normalize the hidden units using L2 normalization, which constrains the representation to be on a hypersphere. Why is that helpful or needed?

Get this bounty!!!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.