Encog3 is a powerful tool for building neural nets in C# and Java very quickly and easily. If you are looking to get into AI, or if you need to use a neural net in your project without the enormous overhead of rolling your own, you will want to get to know Encog. This book is a thorough introduction to the tool. How to set it up. How to use it. How it does, what it does. It's all you need to get started with Encog3.
The only thing I did not like was the obvious hand of a project manager in the editing. You know the type. You write a sentence like "This is the house that Jack built" and they read half a sentence and complain "'That Jack built' isn't a sentence. It should be 'Jack built that.'" Ultimately you end up with, "This is Jack's Jack-built house, built by Jack, who built the house that Jack built."
In this book, that takes the form of:
This section will detail how to structure a neural network for a very simple problem: to design a neural network that can function as an XOR operator. Learning the XOR operator is a frequent "first example" when demonstrating the architecture of a new neural network. Just as most new programming languages are first demonstrated with a program that simply displays "Hello World,"neural networks are frequently demonstrated with the XOR operator. Learning the XOR operator is a sort of the "Hello World" application for neural networks.
It's an unfair structure, because the book is brilliant once you get past the endless repetition, and unnecessary information (they explain the Boolean operators "and" and "or", to an audience sophisticated enough to be using neural nets). The engine itself is revolutionary. The code examples are clear and shockingly concise.