bias/normalisation

[audibert-alexandre-fra]

Hello.
I have one remark

• The "bias" term is not mentionned in your paper but it seems to appear when you compare centers with representations. Moreover there is no normalisation in the previous case.

If my previous comment is correct, can you explain this choice. Thanks in advance

[jiequancui]

Hi,

Thanks for your question.

There are no special reasons.
We observed that after the training with dot products between centers and representations, the output has already been almost normalized.

Best,