Summary

This chapter introduced you to DL, a subcategory of ML that leverages artificial neural networks to mimic human brains and perform automated tasks without human intervention. DL has certainly come to the fore in the last few years because of the incredible advancements in the availability of big data, sophisticated algorithms, and improvements in computational hardware such as CPUs and GPUs. We started this chapter by understanding why there is a need for sophisticated algorithms to mine insights from ever-growing genomics data and how DL, using DNNs, can fill that gap. The anatomy of the neural network architecture, along with the key components of neural networks, was introduced. Understanding these key concepts is important to be able to build a solid foundation for DL concepts, as well as understand how they relate to genomic applications. Then, you were introduced to the different neural network architectures, such as CNNs, RNNs, GANs, GNNs, and autoencoders, and understood...