A Review on Deep Neural Network Design and Their Applications

Deep learning has experienced considerable success in many application domains in recent days. Different approaches have been suggested based on different learning groups, including supervised, semi-supervised, and unsupervised learning approaches. In this paper, we address some commonly used design for deep learning and their practical applications are highlighted. There are innumerable architectures available with an recent improvement in proposed algorithms like Convolutional Neural Network (CNN), Radial Basis Function Networks, Multilayer Perceptron's, Self -Organizing Maps, Deep Belief Networks, Restricted Boltzmann Machines, Radial Basis Function Networks, Autoencoders etc. Various models of deep neural networks are investigated and latest advances are summarized. Applications of deep learning techniques in some field's Natural language processing, information retrieval, analysis of social networks, transportation prediction and sound processing are tinted.