3Blue1Brown Neural Networks Series
Visual explanations of what a neural network is, how gradient descent works and what backpropagation is doing, plus later episodes on transformers and attention. The mathematics stops being intimidating when you can see it move.
Video
AI & Machine Learning
Grant Sanderson
About 3 hours
Beginner
Free
What it is
An animated series explaining the mathematics underneath neural networks, and more recently transformers and attention, through visualisation rather than notation.
Why it works
Gradient descent is a shape. Backpropagation is a flow. Attention is a lookup. All three are far easier to hold in your head as animated objects than as equations, and once you have the picture the equations become readable rather than opaque.
How to use it
Watch it before any course that assumes mathematical comfort. Three hours here makes the following twenty hours of any machine learning material substantially easier, particularly if your calculus is rusty.
Best for: anyone starting machine learning who finds the mathematical notation a barrier.
Ready to start?
Opens on Grant Sanderson in a new tab.
Stuck on something specific?
Writing only gets you so far. If you want an answer to your situation rather than the general case, book a session and we will work through it together. Every session is free; a few slots open each week.
Follow along
New writing, resources and project ideas land here first.