Deep Learning - An Emerging Paradigm

Abstract

In the past few years, deep learning has emerged as a vital paradigm in various fields such as vision, semantic understanding, natural language processing and speech recognition. A number of architectures and algorithms have been proposed and used for the extraction of multiple levels of feature abstractions. Deep learning is a great research interest nowadays due to its commendable properties in contrast to conventional neural network which was not so successful to generate favorable outcomes in many applications because of having shallow architecture and handcrafted features. This paper focuses on providing an understanding of different deep learning architectures and emerging application areas of deep learning such as internet of things, big data and blockchain.