YOUR FOUNDATION-FIRST SYLLABUS
Image generation
Start with how computers store pictures. Learn how a model learns from examples and gradually turns random visual noise into a picture guided by a description.
0 of 73 core ideas demonstrated72 lessons940 estimated minutes remaining
Already know some of this?
Go straight to an idea’s knowledge check. Passing its different checks without hints carries that evidence into this path. Checking an advanced idea does not award its prerequisites; you can still explore them.
Check what I already knowTHE IDEAS, IN ORDER
Shared credit is automaticBuild your understanding.
Chapter counts use compatible knowledge-check evidence. Reading a lesson does not mark its ideas as demonstrated.
Chapter 1 · Lessons 1–7 Meet an AI system Start with an everyday task, then follow its inputs, model, and output. 7 lessons · 0 of 7 core ideas demonstrated Up next
01
What is AI? Start with an everyday task Common Ground · 11 min
To explore02 Inputs and outputs: what goes in and what comes out Common Ground · 13 min
To explore03 Data and examples: what a computer can learn from Common Ground · 10 min
To explore04 What is a model? A small rule inside a bigger app Common Ground · 10 min
To explore05 Weights: the adjustable numbers in a model Common Ground · 12 min
To explore06 Training and inference: changing a rule or using it Common Ground · 13 min
To explore07 Follow one small AI system from start to finish Common Ground · 9 min
To exploreChapter 2 · Lessons 8–14 How computers represent a problem Explore everyday AI ideas; how AI got here; how computers represent a problem; and supporting ideas. 7 lessons · 0 of 7 core ideas demonstrated
08
Different ways to solve the same problem Common Ground · 11 min
To explore09 How to tell whether an AI answer is useful Common Ground · 10 min
To explore10 The questions that shaped AI Common Ground · 15 min
To explore11 How a computer stores a picture or a sentence Learning mechanisms · 16 min
To explore12 Describe a computation so someone else can follow it Learning mechanisms · 11 min
To explore13 What a representation makes easy to learn Learning mechanisms · 12 min
To explore14 The numbers you actually need Math preparation · 10 min
To exploreChapter 3 · Lessons 15–21 Making sense of images Explore making sense of images; learning from useful data; learning to predict; and supporting ideas. 7 lessons · 0 of 8 core ideas demonstrated
15
Vectors: lists that work together Math preparation · 10 min
To explore16 Distances, neighborhoods, and transformations Math preparation · 12 min
To explore17 What changes when an image changes? Learning mechanisms · 13 min
To explore18 From the world to a dataset Learning mechanisms · 11 min
To explore19 A function is a rule Math preparation · 9 min
To explore20 Build a tiny prediction model Learning mechanisms · 11 min
To explore21 How does a model improve a prediction? Learning mechanisms · 32 min
To exploreChapter 4 · Lessons 22–29 Generating something new Explore inside a neural network; generating something new; math preparation when it is needed. 8 lessons · 0 of 8 core ideas demonstrated
22
Read an equation one symbol at a time Math preparation · 10 min
To explore23 A derivative is a local change Math preparation · 13 min
To explore24 The chain rule, one step at a time Math preparation · 13 min
To explore25 How an error reaches earlier weights Learning mechanisms · 11 min
To explore26 How an image can emerge from noise Your system, connected · 21 min
To explore27 Probability without the mystery Math preparation · 10 min
To explore28 Predicting a label, generating a possibility Learning mechanisms · 14 min
To explore29 Why image generators work in latent space Your system, connected · 15 min
To exploreChapter 5 · Lessons 30–36 Inside a neural network: Inside an artificial neuron Explore inside a neural network; helping a model improve; math preparation when it is needed. 7 lessons · 0 of 7 core ideas demonstrated
30
Inside an artificial neuron Learning mechanisms · 14 min
To explore31 Why layers need more than multiplication Learning mechanisms · 12 min
To explore32 Follow shapes through a numerical layer Math preparation · 10 min
To explore33 Trace inputs through layers to a loss Learning mechanisms · 17 min
To explore34 Let the graph carry derivatives Learning mechanisms · 12 min
To explore35 Averages over uncertain outcomes Math preparation · 9 min
To explore36 Learn from a sample of examples at each step Learning mechanisms · 11 min
To exploreChapter 6 · Lessons 37–43 Helping a model improve Explore helping a model improve; math preparation when it is needed. 7 lessons · 0 of 7 core ideas demonstrated
37
Several inputs, several rates of change Math preparation · 9 min
To explore38 Choose a direction and a step size Math preparation · 9 min
To explore39 How momentum and Adam use earlier gradients Learning mechanisms · 15 min
To explore40 Constrain a solution or change the update Learning mechanisms · 12 min
To explore41 Powers and logarithms, one step at a time Math preparation · 11 min
To explore42 When the computer cannot store the exact number Math preparation · 12 min
To explore43 Keep signals and gradients in a usable range Learning mechanisms · 12 min
To exploreChapter 7 · Lessons 44–50 Inside a neural network: Help information and learning signals travel Explore inside a neural network; learning from useful data; finding patterns and useful representations; and supporting ideas. 7 lessons · 0 of 7 core ideas demonstrated
44
Help information and learning signals travel Learning mechanisms · 12 min
To explore45 Connect data, gradients, and saved model state Learning mechanisms · 15 min
To explore46 Different ways to learn Common Ground · 12 min
To explore47 Create a learning task from the data itself Learning mechanisms · 14 min
To explore48 What can a network’s hidden features tell us? Learning mechanisms · 12 min
To explore49 From a waveform to time-frequency features Learning mechanisms · 14 min
To explore50 Connect information expressed in different forms Learning mechanisms · 15 min
To exploreChapter 8 · Lessons 51–57 Learning to predict Explore learning to predict; learning from useful data; testing what a model has learned; and supporting ideas. 7 lessons · 0 of 7 core ideas demonstrated
51
Turn a linear score into a class probability Learning mechanisms · 11 min
To explore52 What can a small experiment tell us? Math preparation · 10 min
To explore53 Clean data without erasing the problem Learning mechanisms · 12 min
To explore54 Protect the examples used to judge a model Learning mechanisms · 15 min
To explore55 Control complexity without peeking at the answer Learning mechanisms · 12 min
To explore56 Evaluate the procedure that chose the model Learning mechanisms · 14 min
To explore57 Check whether answers actually follow the supplied media Learning mechanisms · 14 min
To exploreChapter 9 · Lessons 58–64 Working with uncertainty Explore learning from useful data; learning to predict; working with uncertainty; and supporting ideas. 7 lessons · 0 of 7 core ideas demonstrated
58
Who is represented by the data? Learning mechanisms · 14 min
To explore59 Decide what success means before training Learning mechanisms · 12 min
To explore60 Look inside the average score Learning mechanisms · 14 min
To explore61 Probability, likelihood, and what is unknown Learning mechanisms · 11 min
To explore62 Update a belief using new evidence Learning mechanisms · 13 min
To explore63 Value a choice and its later consequences Math preparation · 12 min
To explore64 Turn a probability into a justified action Learning mechanisms · 11 min
To exploreChapter 10 · Lessons 65–72 Checking reliability and behavior Explore testing what a model has learned; how computers represent a problem; checking reliability and behavior; and supporting ideas. 8 lessons · 0 of 8 core ideas demonstrated
65
Change one thing and measure what follows Learning mechanisms · 13 min
To explore66 Repeat an experiment without confusing luck with truth Learning mechanisms · 11 min
To explore67 Make the result possible to inspect Learning mechanisms · 12 min
To explore68 Measure the behavior that the task actually needs Learning mechanisms · 14 min
To explore69 Build a test whose score supports the intended claim Learning mechanisms · 12 min
To explore70 Treat evaluators as fallible measurement instruments Learning mechanisms · 17 min
To explore71 Decide whose problem the system solves and who bears its errors Learning mechanisms · 12 min
To explore72 Project: specify and test a conditional image generator Learning mechanisms · 45 min
To exploreOptional extensions and comparisons (19)
These lessons deepen or compare the reference design. They do not add requirements to this path’s completion.
How to use this school without getting lostOptional orientation: how paths, side lessons, and practice fit together, plus study habits that make learning stick.Practice, progress, and knowing what you understandOptional orientation: use practice, mistakes, and spaced review to find out what you really understand.Misuse in practice: jailbreaks, deepfakes, and provenanceA familiar face or voice is no longer proof of who is speaking. Learn how misuse works and the checks that still hold.Train a generator against a learned discriminatorGANs: two networks competing, the method behind the first photorealistic fake faces.Generate different kinds of data one unit at a timeGenerate images or video one token at a time, the way language models write, and compare that with diffusion.Represent probability through invertible maps or energy scoresTwo more ways to model images: invertible flows and energy scores.Learn a velocity field and integrate it into samplesThe fast successor to diffusion sampling: learn a velocity field and follow it from noise to image.Trace an example through a batchFollow array shapes through a batch: the everyday skill of reading and debugging model code.Why a small choice can create a huge searchWhy trying every possibility explodes, and why language models can't search every sentence.What makes an accelerator useful?Why GPUs made modern AI possible, and why more chips don't always mean more speed.Be a good scientist with a small examplePractice a scientist's habits (counterexamples, exhaustive tests, held-out data) on a model small enough to check completely.Test what the visual model is actually usingShortcuts, adversarial pixels, and occlusion: stress-test what an image model really learned.Learn a latent representation and a way back to dataThe compress-and-reconstruct idea behind latent diffusion's image encoder.Train alignment, generation, and instruction behavior across modalitiesHow text-to-image and text-to-audio systems are trained stage by stage.Understand what training and deployment can reveal about dataCan a model leak its training data? Memorization, extraction, and differential privacy.Find failures hidden by an overall scoreFind the groups a model quietly fails, hidden inside a good average score. The same slice tests work for photos and text.Choose fairness measurements that match the decision contextMeasure whether an image system works equally well for different groups, and why fairness definitions conflict.Project: compare a baseline with a learned predictorBuild a baseline-vs-learned comparison end to end on a tiny dataset before tackling images.Project: build a two-layer network and check its gradientsTrace one full forward and backward pass by hand before training image models.