Giters

airalcorn2 / Deep-Semantic-Similarity-Model

My Keras implementation of the Deep Semantic Similarity Model (DSSM)/Convolutional Latent Semantic Model (CLSM) described here: http://research.microsoft.com/pubs/226585/cikm2014_cdssm_final.pdf.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Shared weights

michellemay opened this issue · comments

commented

Hi, I stumbled upon your implementation of dssm and was wondering the following: should the different weights matrices be shared between query and doc?

commented

See the last couple of paragraphs in section 5.2 of this paper. You can share the weights if you want, but the authors say using separate nets for the query and documents gives better performance.

commented

Indeed, I missed that line. Thanks.