Convolutional Neural Networks

Abstract

A convolutional neural network (CNN) is essentially a neural network that employs the convolution operation (instead of a fully connected layer) as one of its layers. CNNs are an incredibly successful technology that has been applied to problems where in the input data on which predictions are to be made has a known grid-like topology, like a time series (a 1-D grid) or an image (a 2-D grid). CNNs ushered deep learning into modern times, solving one of the most crucial computational problems in the digital era of computer vision. With the popularity of CNNs, a surge in the research for deep learning was witnessed that continues today.

Topics & Concepts