Although simple neural networks, like the one utilized in the preceding section, are extremely powerful for many scenarios, deep neural network architectures have been applied across industries in recent years on various types of data. These more complicated architectures have been used to beat champions at board/video games, drive cars, generate art, transform images, and much more. It almost seems like you can throw anything at these models and they will do something interesting, but they seem to be particularly well-suited for computer vision, speech recognition, textual inference, and other very complicated and hard-to-define tasks.
We are going to introduce deep learning here and run a deep learning model in Go. However, the application and diversity of deep learning models is huge and growing every day.
There are many books and tutorials on the...
