I stopped about a third of the way through, once machine learning cropped up (I'm biased against it, as I'd rather humans use energy to learn and practice than

I stopped about a third of the way through, once machine learning cropped up (I'm biased against it, as I'd rather humans use energy to learn and practice than a computer do it, but I accept I might be missing something), but I support the first two points: 1. Define the problem and why we're going to solve it (in D&D terms, I'd say this is a wisdom check) 2. Translate the problem into math language (needs imagination and experience, which is an intelligence check) 2.5. Levitt noted the importance of interleaved practice; why practice a tool intensively for the test only to never use it again? Supported by the authors of Make it Stick: the science of successful learning. That all sounds better than "here kids, learn these tools out of context".