2 ms·

why would anyone assume deep learning to beat everything at everything, even applied relatively naively?