[Lisa_seminaires] [Tea Talk] Yaroslav Ganin (MILA) Fri June 15 10:30AM AA3195

Michael Noukhovitch mnoukhov at gmail.com
Mar 12 Juin 13:35:53 EDT 2018


This week we have our very own Yaroslav Ganin giving a talk on the work he
did on internship at DeepMind on Friday June 15th at 10:30AM in room AA3195

This talk may be streamed as usual here
<https://bluejeans.com/809027115/webrtc>
And you can sign up to meet the speaker by messaging him on slack

If you don't come to the talk, be prepared for a strongly reinforced
adversary in the form of me!
Michael

*TITLE* Synthesizing Programs for Images using Reinforced Adversarial
Learning

*KEYWORDS *GAN, Reinforcement Learning, Unsupervised Learning

*ABSTRACT *
Advances in deep generative networks have led to impressive results in
recent years. Nevertheless, such models can often waste their capacity on
the minutiae of datasets, presumably due to weak inductive biases in their
decoders. This is where graphics engines may come in handy since they
abstract away low-level details and represent images as high-level
programs. Current methods that combine deep learning and renderers are
limited by hand-crafted likelihood or distance functions, a need for large
amounts of supervision, or difficulties in scaling their inference
algorithms to richer datasets. To mitigate these issues, we present SPIRAL,
an adversarially trained agent that generates a program which is executed
by a graphics engine to interpret and sample images. The goal of this agent
is to fool a discriminator network that distinguishes between real and
rendered data, trained with a distributed reinforcement learning setup
without any supervision. A surprising finding is that using the
discriminator’s output as a reward signal is the key to allow the agent to
make meaningful progress at matching the desired output rendering. To the
best of our knowledge, this is the first demonstration of an end-to-end,
unsupervised and adversarial inverse graphics agent on challenging real
world (MNIST, Omniglot, CelebA) and synthetic 3D datasets
-------------- section suivante --------------
Une pièce jointe HTML a été nettoyée...
URL: http://webmail.iro.umontreal.ca/pipermail/lisa_seminaires/attachments/20180612/70f2a1c6/attachment.html 


Plus d'informations sur la liste de diffusion Lisa_seminaires