[Lisa_seminaires] [TeaTalk] Karan Grewal, September 8, 11:00, AA6214 (TIME CHANGED!)

Dzmitry Bahdanau dimabgv at gmail.com
Jeu 7 Sep 15:57:25 EDT 2017


Hi all,

Our tea-talk tomorrow has been moved to 11am, same room (AA6214). Hope to
you see there!

Dima

On Mon, 4 Sep 2017 at 21:02 Dzmitry Bahdanau <dimabgv at gmail.com> wrote:

> Hi all,
>
> Our next speaker is Karan Grewal, who is currently an intern at MILA. The
> talk will take place at *AA6214, at 13:45, September 8*. Hope to see many
> of you there!
>
> *Title: *Variance Regularizing Adversarial Learning
>
> *Abstract:* Generative Adversarial Networks have become a breakthrough
> towards generating synthetic data in recent years, however many problems
> may arise during training which result in the generator being unable to
> learn the desired data distribution/manifold. Most notably, the
> discriminator often overpowers the generator, which leads to a poor
> training signal and vanishing gradients. We study the effects of the
> discriminator's unnormalized output distribution on the generator's ability
> to learn. To combat this problem, we propose Variance Regularizing
> Adversarial Learning -- regularizing the discriminator's unnormalized
> output distribution to fit a mixture of gaussians, yielding higher variance
> and hence a lipschitz-like discriminator function. To do this, we provide
> two methods: (1) using the KL-divergence as a penalty in the
> discriminator's loss and (2) playing meta-adversarial games to force the
> discriminator to fit a MoG. We show that our models are robust in the
> presence of high training ratios and compare them to other variants of GANs.
>
> *Bio:* Karan Grewal is an intern at MILA and a senior undergraduate
> student at the University of Toronto. He is working on improving and
> stabilizing generative models with Devon Hjelm and Yoshua Bengio.
> Previously, he worked on applying machine learning to social contexts.
>
> Dima
>
-------------- section suivante --------------
Une pièce jointe HTML a été nettoyée...
URL: http://webmail.iro.umontreal.ca/pipermail/lisa_seminaires/attachments/20170907/8b4e8d6b/attachment.html 


Plus d'informations sur la liste de diffusion Lisa_seminaires