Save this video free

Lecture 8: Generative Adversarial Networks (KAIST CS479, Fall 2026)

Minhyuk Sung · 1:03:00 · Watch on YouTube

Part of KAIST AIC313/CS378. See all 12 lectures and search them

These notes were made with YouTube Collector. Paste any YouTube link and get the same: a summary, chapters and a searchable transcript.

Overview

Minhyuk Sung compares autoregressive models, normalizing flows, and VAEs before explaining how Generative Adversarial Networks (GANs) learn through competition between a generator and a real-versus-fake discriminator, rather than by directly maximizing likelihood. He derives the binary cross-entropy objective, connects the optimal discriminator to the Jensen-Shannon divergence, and covers practical challenges including vanishing gradients, unstable minimax training, and mode collapse.

Key takeaways

Chapters

0:00 Autoregressive Models, Normalizing Flows, and VAEs: Different Tradeoffs
5:00 Jensen-Shannon Divergence: A Symmetric Distribution Comparison
9:34 GANs Generate Samples Without Computing Likelihood
11:59 Deep Learning's Rise and the Reliability Question
15:59 Adversarial Examples Reveal Classifier Vulnerability
19:59 GAN Architecture: A Generator Competes With a Discriminator
23:18 Binary Cross-Entropy Defines the Discriminator Objective
30:52 The Optimal Discriminator Estimates a Density Ratio
40:25 An Optimal Discriminator Connects GAN Training to JS Divergence
45:03 Alternating Updates Require Freezing Network Parameters
49:06 Non-Saturating Generator Loss Addresses Vanishing Gradients
55:32 GAN Quality, Mode Collapse, and the Diffusion Tradeoff

Keep these chapters and the full searchable transcript in your own library.

Transcript excerpt

0:02 Yeah. So basically this quiz shows like multiple things. So basically as we discussed last time with this kind of the gausian the base the one layer of the offline the model can be actually interpreted especially the gaus and the typical a model right so in the sense it sounds like normalizing flow seems kind of like something that's like better than the or model because now we can stack multiple these kind of the flow layers in terms like while basically permuting the order the variables right so in terms of basically permitting the order of the variables. Now we can increase the expressiveness of the D model. Right? So it sounds like the normalizing flow is kind of the better than the original model. But actually that's not exactly the case because this

0:49 is the the case that we only compare the normalizing flow with the Gaussian base the ARD model for the orderless model. We can basically model the conditional distribution not as kind of the single gausian but also with the gausian the mixture and also other types of the distribution. It can be the binary distribution and also the the categorical distribution right. So the good thing of the model is that we can also easily handle like all the discrete data as well. So that's kind of also the difference between the the model and the normalized flow. Right? So we are seeing some kind of the uh some kind of some common things and also kind of differences between these two cases and also when you compare the the normalizing flow with the BA also the normalizing flow might look

1:37 kind of be better in terms of that that's basically the model that allow us to compute the exact likelihood. So like

This is the start of the transcript. Save this video free for the whole transcript, searchable and time-coded.

In this course

Related notes

Summary, takeaways, and chapters were generated by AI from the video's transcript and may contain errors. The video belongs to its creator, Minhyuk Sung.

Want the full transcript?

Save this video in YouTube Collector to get its complete searchable transcript, your own AI summaries, and a library that keeps every video you collect in one place.