likhuva

In my opinion, the fundamental explanations you seek lie in Probability Theory, not matrix theory. When it comes to ML, matrices are just implementation details. I highly suggest this set of notes: https://chrispiech.github.io/probabilityForComputerScientist...

For Stanford, you can start here: https://explorecourses.stanford.edu/.This should give you enough information about the various classes and their professors/lecturers. From there you can try searching for the course numbers on google. Most will have a website that hosts most of the material you may be interested in.

>> no compilers or os, no multivariable calculus, no pointers.

Stanford still requires cs students to take classes that cover the material above (except compilers which can be taken as a specialization)

>> seems the downvoters don’t like what I have to say, but that doesn’t make it less true.

Except that it isn't true. A quick look at https://cs.stanford.edu/degrees/ug/Requirements.shtml paints a completely diffrent picture.