If you just have a single problem to solve then fine go ahead and use a neural network. But if you want to do science and understand how to choose architectures or how to go to a new problem you have to understand what different architectures can and cannot do.
The thing that interests me least about the radio business is the radio business. But I've had to learn a little bit about it. It's not rocket science: You get ratings that's good.
Next Article
Weapons of mass destruction aren't pulled out of a black hat like a white rabbit at a magic show. They're produced in factories. There's science and technology involved. They're not produced in a hole in the ground or in a basement.
