Review of deep learning network

Abstract

∗Deep learning is a technology that uses the hierarchical structure of neural network to learn features. It allows computer models with multiple processing layers to learn and represent data like the brain’s perception and understanding of multimodal information, so as to implicitly capture complex large-scale data. The whole system of deep learning network forms a hierarchical and powerful feature representation structure, which enables it to analyze and extract useful knowledge from a large amount of data. This paper mainly introduces the development and application of supervised convolution neural network, unsupervised convolution neural network and generative countermeasure network, and analyzes the research status and challenges of deep learning network. Through the review and introduction of important papers on deep learning network, it provides researchers with accessible scientific research materials.