Save this video free

CMU Introduction To Deep Learning 11-785, Fall 2026: Lecture 23

Carnegie Mellon University Deep Learning · 1:21:46 · Watch on YouTube

Part of Carnegie Mellon University 11-785 Introduction to Deep Learning. See all 24 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

The lecture develops diffusion models as a way to generate data by learning to reverse a fixed Gaussian-noising process, using a simple noise-prediction MSE objective rather than directly modeling the full data distribution. It connects DDPM training to score matching and reverse SDEs, explains how DDIM accelerates sampling, and covers classifier-free guidance, evaluation metrics, latent diffusion, ControlNet, DreamBooth, and video generation.

Key takeaways

Chapters

0:00 Lecture Roadmap: From Diffusion to Applications
3:18 Generative Models Learn Data Distributions, Not Labels
8:07 Why Diffusion Uses Many Small Denoising Steps
13:13 Brownian Motion and the DDPM Noise-Prediction Insight
16:15 Gaussian Forward Diffusion and the Reparameterization Shortcut
22:41 Learning the Reverse Process with Noise-Prediction MSE
26:28 Noise Schedules and the DDPM Training Loop
32:34 A 2D Spiral Shows How Denoising Changes with Noise Level
40:19 Optimal Denoising, Data Manifolds, and Score Geometry
48:58 DDPM Sampling and DDIM's Faster Non-Markovian Updates
56:42 Continuous-Time Diffusion: ODEs and Reverse SDEs
1:02:07 Score Matching Connects Noise Prediction to Probability Gradients
1:07:00 Classifier-Free Guidance Steers Conditional Generation
1:13:21 Evaluating Generative Models with FID, IS, Precision, and Recall
1:14:48 Latent Diffusion and Stable Diffusion's VAE Sandwich
1:17:42 ControlNet, DreamBooth, Video Diffusion, and Core Lessons

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

Transcript excerpt

1:49 I want you to stand with guidance on >> [snorts] [clears throat] >> Okay. So let's start.

3:22 Good morning everyone. Today we're going to explore one of the most exciting developments in generative AI which is diffusion model. So here is our roadmap for today. So we'll start by defining what's diffusion. We will explain the the process step by step. We'll look at the generation and the inference. What is DDIM? Then after that we will start exploring SDEs and score matching. We will also be talking about a classifier free guidance which gives us control on what to generate. Finally we will be um covering the evaluation and the real world applications. So, [snorts] to understand diffusion, we first have to ask, what's the goal of

4:11 the model? So, we have discriminative models [clears throat] which actually learn a mapping from an input to a label Y from input X to label Y. Essentially, we're learning a decision boundary. So, you can ask, is this a cat or a dog? But generative models, they're doing something much harder. They're not looking for a boundary. They're trying to learn the entire probability distribution of the data. So, they want to understand the essence of a cat so well that they can draw a brand new one from scratch. So, technically, we're moving from computing P Y given X to modeling the data distribution which is PX.

4:59 So, what does it mean to learn to generate data? So, how do we learn a distribution? Imagine every cat in our training set is a point in high dimensional space. These points, they cluster together on what we call the data manifold. So, the goal of generative um model is to take a simple, easy to sample distribution like Gaussian noise and

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, Carnegie Mellon University Deep Learning.

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.