MedGAN
b-GAN
LSGAN
ID-CGAN
LS-GAN
AffGAN
CoGAN
LAPGAN
LR-GAN
DiscoGAN
AMGAN
CGAN
MPM-GAN
iGAN
IcGAN
AdaGAN
IAN
InfoGAN
Generative Adversarial Networks
CatGAN
Ian Goodfellow, Staff Research Scientist, Google Brain
alpha-GAN
FF-GAN
C-RNN-GAN
ICCV Tutorial on GANs
Venice, 2017-10-22
GoGAN
GMAN
MIX+GAN
BS-GAN
DR-GAN
AC-GAN
DCGAN
McGAN
MGAN
C-VAE-GAN
MAGAN
GAWWN
CCGAN
3D-GAN
DualGAN
Bayesian GAN
EBGAN
ALI
MARTA-GAN
WGAN-GP
Context-RNN-GAN
f-GAN
ArtGAN
CycleGAN
AnoGAN
MAD-GAN
BEGAN
MalGAN
BiGAN
GP-GAN
DTN
AL-CGAN
Generative Modeling
• Density estimation
• Sample generation
Training examples
Model samples
(Goodfellow 2017)
Maximum Likelihood
BRIEF ARTICLE
THE AUTHOR
✓⇤ = arg max
✓
Ex⇠pdata log pmodel(x | ✓)
(Goodfellow 2017)
Adversarial Nets Framework
D(x) tries to be
near 1
Differentiable
function D
D tries to make
D(G(z)) near 0,
G tries to make
D(G(z)) near 1
D
x sampled from
data
x sampled from
model
(Goodfellow et al., 2014)
Differentiable
function G
Input noise z
(Goodfellow 2017)
What can you do with GANs?
• Simulated environments and training data
• Missing data
• Semi-supervised learning
• Multiple correct answers
• Realistic generation tasks
• Simulation by prediction
• Solve inference problems
• Learn useful embeddings
(Goodfellow 2017)
(Goodfellow 2017)
GANs for simulated training data
(Shrivastava et al., 2016)
(Goodfellow 2017)
GANs for domain adaptation
(Bousmalis et al., 2016)
(Raffel, 2017)