AI/ML from the Ground Up

a from-scratch explainer

AI/ML from the Ground Up

The whole field, built from first principles, until you can pick the right model for any problem.

Start with Chapter 1 →

Hi, I’m Sean Egan, a Navy information warfare officer and research scientist. I built this so you can look at a fresh problem and say, with reasons, which model it calls for.

Every idea starts with a picture, then the math in plain English, then something to try yourself. The LLM book has the unhurried proofs. Start at Chapter 1, or Chapter 3 if you already know vectors.

(The trailer starts muted — tap the speaker icon for narration. Chapter videos play with sound.)

What you will be able to do · narrated, with captions.

Part I · What Learning Means

  1. Learning from DataLearning rules from examples.10 min
  2. The Math ToolkitArrows, slopes, and chance.12 min
  3. Fitting a LineLearning by reducing error.12 min
  4. Drawing BoundariesDeciding which class fits.11 min

Part II · Neural Networks

  1. The NeuronStacking small computing units.12 min
  2. Teaching the NetworkTraining without memorizing.11 min
  3. The Classical ToolboxChoosing tools for tables.12 min

Part III · Architectures for Different Worlds

  1. Seeing MachinesFinding patterns in pictures.13 min
  2. Remembering MachinesCarrying information through sequences.13 min
  3. Networks of RelationshipsLearning by passing messages.10 min
  4. Making Things UpLearning to generate data.10 min
  5. Learning by DoingLearning from actions and rewards.12 min

Part IV · Mastery

  1. Choosing Your ToolMatching models to problems.11 min

One of three books: How LLMs Actually Work · this book · Diffusion (images).