Deep learning models
In this section, we will dive into three popular deep learning models one by one: CNNs, Restricted Boltzmann Machines (RBM), and the recurrent neural network (RNN).
Convolutional Neural Networks
Convolutional Neural Networks are biologically-inspired variants of the multilayer perceptron and have been proven to be very effective in areas such as image recognition and classification. ConvNets have been successfully applied when identifying faces, objects, and traffic signs as well as powering vision in robots and self-driving cars. CNNs exploit spatially-local correlation by enforcing a local connectivity pattern between neurons of adjacent layers. In other words, the inputs of hidden units in the layer m are from a subset of units in layer
, units that have spatially contiguous receptive fields.
LeNet was one of the very first convolutional neural networks proposed by Yann LeCun in 1988. It was mainly used for character recognition tasks such as reading zip codes, digits...
