3 ms·
This is one of the silliest things I have ever read in Slate, and as it is Slate we're talking about: that's saying a lot. But hey, what do I know, I only do ma
by scottlocklin 11y ago
This is one of the silliest things I have ever read in Slate, and as it is Slate we're talking about: that's saying a lot. But hey, what do I know, I only do machine learning for a living... Presumably the knucklehead who wrote this finds linear regression and matrix decomposition to be extremely mysterious....
- gnodar 11y agoIt would perhaps help those of us who know little of machine learning, if you could give specific examples of silly things stated in the article.
- scottlocklin 11y agoStarting from the second sentence: "there’s a burgeoning, alternative model of programming and computation that sidesteps the limitations of the classic model, embracing uncertainty, variability, self-correction, and overall messiness. It’s called machine learning" All the way through the end: there is nothing there which is even remotely within the realm of being even partially correct. It's sort of like when Gladwell yammered on about "Igon values" in a New Yorker article, except worse.