Summary

We have now come to the end of this chapter. We have looked at several variants of artificial neural networks, including CNNs for image processing purposes and RNNs for natural language processing purposes. Additionally, we looked at RBMs and GANs as generative models and autoencoders as unsupervised methods that cater to a lot of problems, such as noise reduction or deciphering the internal structure of the data. Also, we touched upon reinforcement learning, which has made a big impact on robotics and AI.

You should now be familiar with the core techniques that we are going to use when building smart AI applications throughout the rest of the chapters in this book. While building the applications, we will take small technical digressions when required. Readers that are new to deep learning are advised to explore more about the core technologies touched upon in this chapter for a more thorough understanding.

In subsequent chapters, we will discuss practical AI projects, and we will implement them using the technologies discussed in this chapter. In Chapter 2, Transfer Learning, we will start by implementing a healthcare application for medical image analysis using transfer learning. We hope that you look forward to your participation.