Introducing Convolutional Neural Networks for Genomics

In recent years, deep learning (DL) has emerged as a prominent technology in solving complex problems in various domains. Among DL algorithms, convolutional neural networks (CNNs) dominate the current DL applications because of their incredible accuracy in computer vision (CV) and natural language processing (NLP) tasks. A CNN is a type of neural network (NN) architecture that is used for unstructured data and was originally designed to fully automate the classification of handcrafted characters. Some popular applications of CNNs include facial recognition, object detection, self-driving cars, auto-translation, handwritten character recognition, X-ray image analysis, cancer detection, biometric authentication, and so on. Compared to feed-forward NNs (FNNs), which we learned about in the previous chapter, CNNs process multiple arrays using convolutions within a local field, like perceiving images by eye. Thanks to next-generation...