In this chapter, we discussed how to recognize patterns in images in order to distinguish between different classes by covering fundamental principles of deep learning and discussing how to implement them with the deeplearning4j library. We started by refreshing the basic neural network structure and discussed how to implement them to solve handwritten digit recognition problem.

In the next chapter, we'll look further into patterns; however, instead of patterns in images, we'll tackle patterns with temporal dependencies that can be found in sensor data.