[falcor84]

But how does it get good at "computing"? The way I see it, we either program them to do so manually, or we use ML, at which case the model "fits" the computation based on training examples or environmental feedback, no? What am I missing?

[ithkuil]

the distinction is fuzzy indeed, especially if any thing that you "program in manually" has some parameters that are learned.

Conceptually we already have parts of the model that are not learned: the architecture of the model itself.