Deep Learning: A Brief Study on Its Architectures and Applications

Abstract

Deep learning (DL) is a specialized branch of machine learning (ML) and artificial intelligence (AI) that uses multilayered neural networks, and teaches computers to process data in a way inspired by the functionality of human brain. It is based on “deep” neural networks comprising millions to billions of parameters organized into hierarchical layers that mimic human brain. It is a set of methods that uses deep architectures to learn high-level feature representations that allows the learning of more complex models compared to shallow architectures. Deep neural networks are a type of artificial neural network that have many layers in between (deep) the input and output layer. The DL has become one of the most popular and visible areas of ML due to its success in a variety of applications, such as computer vision, natural language processing, and reinforcement learning. Recently, it has become increasingly popular due to the advances in processing power and the availability of large datasets. The aim of this review is to provide an overview on DL frameworks in briefly.