Lecture 20: Generative Models II

Lecture 20 continues our discussion of generative models. We continue our discussion of variational autoencoders, and see how their latent space attempts to disentangle factors of variation in the data. We see how variational autoencoders can be used to sample new data, or to edit existing data. We briefly see how autoregressive and variational models can be combined, as in the VQ-VAE2 architecture. We then discuss generative adversarial networks (GANs) as a class of generative models that do not explicitly model probability densities. We show that at optimality, the distribution modeled by the generator in a GAN minimizes the Jensen-Shannon divergence (JSD) with the true data distribution. We discuss common GAN architectures including DC-GAN, WGAN, StyleGAN, and BigGAN. We see how GANs can be used for conditional generation tasks including class-conditional generation, text-to-image, image super-resolution, and image-to-image translation.

Slides: http://myumi.ch/wlnXq
_________________________________________________________________________________________________

Computer Vision has become ubiquitous in our society, with applications in search, image understanding, apps, mapping, medicine, drones, and self-driving cars. Core to many of these applications are visual recognition tasks such as image classification and object detection. Recent developments in neural network approaches have greatly advanced the performance of these state-of-the-art visual recognition systems. This course is a deep dive into details of neural-network based deep learning methods for computer vision. During this course, students will learn to implement, train and debug their own neural networks and gain a detailed understanding of cutting-edge research in computer vision. We will cover learning algorithms, neural network architectures, and practical engineering tricks for training and fine-tuning networks for visual recognition tasks.

Course Website: http://myumi.ch/Bo9Ng

Instructor: Justin Johnson http://myumi.ch/QA8Pg