| 1 |
September, 2 |
Logistics. Motivation. Autoregressive models (MADE, WaveNet, PixelCNN). |
slides |
video |
| 2 |
September, 9 |
Bayesian framework. Latent variable models. EM-algorithm. |
slides |
video |
| 3 |
September, 16 |
EM-algorithm. VAE. Mean field approximation. |
slides |
video |
| 4 |
September, 23 |
Flow models (NICE, RealNVP, RevNet, i-RevNet). |
slides |
video |
| 5 |
September, 30 |
Flow models (Glow, Flow++). Flows in VAE. Autoregressive flows (IAF). |
slides |
video |
| 6 |
October, 7 |
Autoregressive flows (IAF, MAF, Parallel WaveNet). ELBO surgery. |
slides |
video |
| 7 |
October, 14 |
VampPrior. Posterior collapse (PixelVAE, VLAE). Decoder weakening. IWAE. |
slides |
video |
| 8 |
October, 21 |
Vanila GAN. Vanishing gradients, mode collapse. KL vs JSD. DCGAN. Wasserstein distance. |
slides |
video |
| 9 |
October, 28 |
Wasserstein GAN. Spectral Normalization GAN. f-divergence. |
slides |
video |
| 10 |
November, 11 |
GAN evaluation. Advanced GANs (SAGAN, BigGAN, ProGAN, StyleGAN). |
slides |
video |
| 11 |
November, 25 |
Disentanglement (InfoGAN, beta-VAE, DIP-VAE, FactorVAE). |
slides |
video |
| 12 |
December, 9 |
Continious-in-time models (NeuralODE, FFjord). Quantized latent models (VQ-VAE, VQ-VAE-2, FQ-GAN). |
slides |
video |