In the last chapter, we learned about what meta learning is and different types of meta learning techniques. We also saw how to learn gradient descent by gradient descent and optimization as a model for few-shot learning. In this chapter, we will learn one of the most commonly used metric-based one-shot learning algorithms called siamese networks. We will see how siamese networks learn from very few data points and how they are used to solve the low data problem. After that we will explore the architecture of siamese networks in detail and we will see some of the applications of siamese networks. At the end of this chapter, we will learn how to build face and audio recognition models using siamese networks.

In this chapter, you will learn the following:

• What are siamese networks?
• Architecture of siamese networks
• Applications of siamese networks
• Face recognition using siamese networks
• Building an audio recognition model using siamese networks