[Lisa_teatalk] TeaTalk: Discriminative Regularizers and Metrics for Generative Models

Jörg Bornschein bornj at iro.umontreal.ca
Mon Jan 18 23:27:12 EST 2016


Hi,

This Friday Alex Lamb will talk about ongoing work and cool new results
related to generative modeling:

When: Friday, 22th Jan, 14:30-15:30
Where: Z-209 Pavilion Mc Nicoll
Who: Alex Lamb
Title: Discriminative Regularizers and Metrics for Generative Models

DeepDream and DeepStyle have shown that the representations learned by
discriminatively trained classifiers can be used to produce photorealistic
transformations of images.  In this talk, we explore the question of
whether these discriminatively learned representations can be used to
enhance the quality of generative models.  Our conjecture is that labels
correspond to characteristics of natural data which are the most salient to
humans: identity in faces, objects in images, and utterances in speech.  We
explore the use of this idea as both a regularizer and an evaluation metric
for generative models and present preliminary experimental results on SVHN
and CIFAR.


See you there,

   j
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://webmail.iro.umontreal.ca/pipermail/lisa_teatalk/attachments/20160119/bf123935/attachment.html 


More information about the Lisa_teatalk mailing list