Thanks, I looked. It is a very good overview. I think this really is the best way. You need to understand at the “big picture” level first. So many other just into the weeds and write code and equations and you don’t see why. Knowing why something is done is important. Learning how to do it comes later.
The comment about needing Linear Algebra is true, but not just for AI and machine learning. Any robot needs it. Linear Algabra is foundational for all of robotics. Look at a Kalman filter used for localization. And then any kind of coordinate transform to get (say) from pixels to real-world or robot-centered x,y,z is using linear algebra. You need at least a good intuition of how it works,
The best online learning for Linear Algebra I found is Khan Academy. It is covered as a slightly simplified version of a lower-division university course. A pretty smart high schooler might be able to follow it. (I have some experience teaching high school)
I found I had forgotten almost everything about Linear Algebra and did the Khan class, end to end. It might be overkill because, as a robot-person, you will never need to know all the proofs. Engineers say “just show me the equations”
You do need an intuitive grasp of linear algebra because, without it you will not even know how to ask Claude to write code for you. Kalman filters, coordinate transformation, robot kinematics, and all of modern AI are all based on linear algebra.