\( \newcommand{\vect}[1]{\mathbf{#1}} \newcommand{\mat}[1]{\mathbf{#1}} \newcommand{\R}{\mathbb{R}} \newcommand{\E}{\mathbb{E}} \newcommand{\Prob}{\mathbb{P}} \newcommand{\T}{^{\top}} \newcommand{\norm}[1]{\lVert #1 \rVert} \newcommand{\abs}[1]{\lvert #1 \rvert} \newcommand{\ihat}{\mkern1mu\hat{\boldsymbol{\imath}}\mkern2mu} \newcommand{\jhat}{\mkern1mu\hat{\boldsymbol{\jmath}}\mkern2mu} \newcommand{\pd}[2]{\frac{\partial #1}{\partial #2}} \newcommand{\argmax}{\operatorname*{arg\,max}} \newcommand{\argmin}{\operatorname*{arg\,min}} \newcommand{\cdata}[1]{\class{c-data}{#1}} \newcommand{\cmodel}[1]{\class{c-model}{#1}} \newcommand{\ctruth}[1]{\class{c-truth}{#1}} \newcommand{\cgrad}[1]{\class{c-grad}{#1}} \newcommand{\closs}[1]{\class{c-loss}{#1}} \newcommand{\cparam}[1]{\class{c-param}{#1}} \newcommand{\csecond}[1]{\class{c-second}{#1}} \renewcommand{\,}{\mkern3mu} \)

Start learning

Where to begin with Applied AI Mastery: free courses on linear algebra and machine learning, a first module of reinforcement learning, what you need before each, and how to study so that it sticks.

Start with one of these four modules, whichever matches where you are. Each is four lessons of about 25 minutes, and the first three open a whole course that you can follow module by module. Everything else that is live is listed in the curriculum.

Pick a starting point

1. Numbers, algebra and notation M01.1 L0 Foundation

   Start here if your school maths is rusty, or you never liked it. Numbers on a line, the rules of algebra as pictures of areas, equations solved on a balance, and a machine-learning formula read aloud symbol by symbol. It opens Math Foundations. You need: arithmetic with whole numbers, fractions and decimals.

2. Vectors as arrows and as data M02.1 L1 Core

   Start here if you want the mathematics under machine learning from the ground up. A vector is an arrow and a list of numbers at once; you will move between the two pictures and see houses, images and words become vectors. It opens Linear Algebra I. You need: school algebra and points on a grid.

3. What learning from data means ML01.1 L1 Core

   Start here if you want to know what a model is and how it learns. You will build a spam filter small enough to check with pencil and paper, first by hand and then from examples. It opens Machine Learning Foundations. You need: school algebra and a little Python.

4. Bellman equations RL01.6 L2 Intermediate

   Start here if you are heading for reinforcement learning. You will derive the equations that every RL algorithm rests on, solve a small problem by hand, and prove why repeating one simple update settles on the answer when future rewards are discounted. You need: probabilities and expectations, sums, and a little linear algebra.

Not sure of your level? Every course carries one, from L0 Foundation to L4 Expert. The curriculum says what each level assumes.

How to study a lesson

The lessons are built around what the research on learning supports. A few habits make them work.

1. Read the question, then guess. Before you read on, say what you think the answer is. A wrong guess is fine. It gives the lesson something to correct.
2. Look at the picture before the equation. Every figure is drawn to carry one idea. Find that idea first. The symbols then have something to attach to.
3. Do the worked example on paper. Cover the solution, try it, then compare line by line.
4. Run the code and change it. Each lesson’s code runs as written. Break it on purpose and see what happens.
5. Answer before you open an answer. The recall questions only work if you try first. Testing yourself and spreading practice over time are the two study techniques that a review of ten of them rated highest. Rereading and highlighting rated low.
6. Come back tomorrow. Each lesson opens with two questions from the one before. Use them.

Keeping track

Press Mark this lesson as done at the end of a lesson and it is ticked wherever its code appears. Your progress is stored in your own browser, nowhere else. There are no accounts and no tracking on this site.

What comes next

Four courses are complete: Math Foundations, Linear Algebra I, Calculus I and Machine Learning Foundations. Python from Zero and Exploratory Data Analysis come next, and then the rest of Ring 1, eight foundation courses in all. The roadmap has the order, and the news page announces each release.