Skip to content
Dictionary / Convolutional neural network

Convolutional neural network

Last updated August 14, 2023

What is a convolutional neural network?

A convolutional neural network (CNN) is a type of artificial neural network that is commonly used in computer vision and image recognition. It is designed to process data that has a grid-like structure, such as an image, and learn to identify patterns and features in the data.

How are CNNs constructed?

A CNN consists of several layers of interconnected nodes, known as neurons. Each layer processes the data in a different way, and the output of one layer becomes the input for the next layer.

  • The first layer, known as the input layer, receives the raw data (e.g. the pixels of an image).
  • The intermediate layers, known as hidden layers, apply various transformations to the data, such as convolution, pooling, and nonlinear activation.
  • The final layer, known as the output layer, generates the predicted result (e.g. the object or scene that is present in the image).

What are the key characteristics of CNNs?

One of the key features of a CNN is its use of convolutional layers. In a convolution layer, the neurons are arranged in a grid, and each neuron is connected to a small region of the input data (e.g. a 3x3 or 5x5 grid of pixels in an image). The convolutional layer applies a filter to the input data, which is a set of weights that determine how the input data should be combined to generate the output. By applying multiple filters at different positions in the input data, the convolutional layer can learn to identify different patterns and features in the data.