Learn how to create and use neural networks in your Java programs. This course teaches you not only how to implement machine learning AI with your own artificial neural networks (ANNs), but also the principles of how artificial neural networks work — to the point that you can implement your own.

You'll need only a knowledge of Java programming and basic algebra; in this course you'll learn the relevant linear algebra, information theory and calculus, and together we'll build a fast and efficient neural network from scratch, able to recognise handwritten digits and easily adapted to other tasks.

Among other things, we’ll cover:

• What artificial neural networks are and how to write them yourself

• How matrixes and linear algebra can be used to create efficient neural networks

• The basic principles of the calculus needed to train your networks

• Writing and organising fast, efficient, multithreaded neural network code

• The fundamental information theory concepts that can enable us to evaluate our neural network performance

• Training your network on the freely-available MNIST hand-written digit database

After taking the course, artificial neural networks won't be a mystery to you any more. You'll be able to write your own neural networks and integrate them seamlessly into your Java programs, and understand in detail how they work.

Whether you’re completely new to neural networks and the relevant mathematics, or you’re using neural network libraries and you know some mathematics but you just don’t know how it all actually works and fits together, this course aims to clear up all the mystery.

Artificial intelligence is an increasingly important technology in the modern world, and this course will teach you the fundamentals of perhaps the most important building block of it.