Notes about convolutional neural networks

by

Let’s start with Wikipedia: In deep learning, a convolutional neural network (CNN, or ConvNet) is a class of deep neural network, most commonly applied to analyze visual imagery.

The name “convolutional neural network” indicates that the network employs a mathematical operation called convolution. The convolutional networks are a specialized type of neural networks that use convolution in place of general matrix multiplication in at least one of their layers.

Representation of the matrix multiplication in one layer:

matrix multiplication in one layer for a convolution neural network
matrix multiplication in one layer for a convolution neural network

A convolutional neural network processes data that has a grid-like arrangement then extracts important features. One huge advantage of using CNNs is that you don’t need to do a lot of pre-processing on images.

A vanilla Convolutional Neural Network (CNN) representation.
A vanilla Convolutional Neural Network (CNN) representation

A basic clasiffication of CNNs is by number of dimensions:

  • 1 Dimension: the CNN kernel moves in one direction. 1D CNNs are usually used on time-series data.
  • 2 Dimensions: the CNN kernel moves in two directions. You’ll see these used with image labelling and processing. These are the more common ones.
  • 3 Dimensions: the kernel moves in three directions. These CNNs are used on images as CT scans and MRIs.

Some links to remind

Leave a Comment