Deep learning and Its Development

Abstract Deep learning is a promising branch of machine learning. It an algorithm that uses artificial neural networks as the architecture to characterize and learn data. In recent years, many companies, for example, Google, Microsoft Baidu, have become interested in field deep set up large-scale projects, such Google’s Deepmind project, including alphago, which has achieved success Go e-sports. This article analyzes summarizes each current research direction approach learning, with prospection about future development expounded. An overview three basic models given, namely multilayer perceptrons perceptrons, convolutional recurrent networks. The benefits superiority are illustrated compared conventional methodology used common applications. Further on emerging types introduced. application various fields summarized, intelligence, computer vision natural language processing applications, some open problems also analyzed. Finally, significance purpose discussed.