Главная
Study mode:
on
1
Intro
2
Which face is fake?
3
Supervised vs unsupervised learning
4
Why generative models? Outlier detection
5
Latent variable models
6
What is a latent variable?
7
Autoencoders: background
8
Dimensionality of latent space → reconstruction quality
9
Autoencoders for representation learning
10
VAEs: key difference with traditional autoencoder
11
VAE optimization
12
Priors on the latent distribution
13
VAEs computation graph
14
Reparametrizing the sampling layer
15
VAEs: Latent perturbation
16
VAE summary
17
Generative Adversarial Networks (GANs)
18
Intuition behind GANS
19
Progressive growing of GANS (NVIDIA)
20
Style-based generator: results
21
Style-based transfer: results
22
CycleGAN: domain transformation
23
Deep Generative Modeling Summary
Description:
Explore deep generative modeling in this lecture from MIT's Introduction to Deep Learning course. Learn about supervised vs unsupervised learning, outlier detection, and latent variable models. Dive into autoencoders, Variational Autoencoders (VAEs), and their optimization techniques. Discover Generative Adversarial Networks (GANs), including progressive growing of GANs and style-based generators. Examine domain transformation with CycleGAN and gain a comprehensive understanding of deep generative modeling techniques and applications.

Deep Generative Modeling

Alexander Amini
Add to list
#Computer Science #Deep Learning #Autoencoders #Machine Learning #Representation Learning #Artificial Intelligence #Generative AI #Generative Models #Data Science #Data Analysis #Anomaly Detection #Outlier Detection #Mathematics #Statistics & Probability #Latent Variable Models
0:00 / 0:00
1 of 23

Intro