Probabilistic Siamese Networks for Learning Representations
We explore the training of deep neural networks to produce vector representations using weakly labelled information in the form of binary similarity labels for pairs of training images. Previous methods such as siamese networks, IMAX and others, have used fixed cost functions such as $L_1$, $L_2$-no...
Main Author: | |
---|---|
Other Authors: | |
Language: | en_ca |
Published: |
2013
|
Subjects: | |
Online Access: | http://hdl.handle.net/1807/43097 |