Deep Stacking Network (DSN) is a deep architecture designed to enable large CPU clusters and benefit from the deep learning capabilities of deep neural networks. The deep stacking network architecture was originally introduced by Li Deng and Dong Yu in 2011 and is commonly referred to as the Deep Convex Network (DCN) to emphasize the depth of the algorithm used to learn the network.

The deep network model has been shown to be a promising paradigm that provides complex data for high-performance machine learning, with a plurality focused on neural networks. More experiments have been performed in recent years with the goal of constructing different deep architectures with the use of non-neural network approaches. The Deep Stacking Network (DSN) model uses easy-to-learn blocks to create a complex parallel parameter training network.

#neural-networks #deep-stacking-network #autoencoder #machine-learning