ЦРУшный подонок Дин:
You could then train it
with some supervised data, saying this is a car, or a truck, or whatever, and get better results than if you just did the unsupervised training.
It was eye-opening for us. It showed that scaling these kinds of models to larger sizes would produce good results, and it started a whole decade of successful uses, at Google and elsewhere, of neural networks for a wide variety of things —
speech recognition, images, language understanding, and so on — as well as some applications in healthcare and medical diagnostics.
https://www.medscape.com/viewarticle/954956#vp_3Этой падле положили на стол мои патентные заявки, через час после того как я подал на них. А через два часа сучьи потроха из Гугля начали делать на них деньги, внаглую воруя.