What is a Convolutional Neural Network (CNN) and why is it used instead of a fully connected network?

A CNN is a neural network designed for grid-like data (images) that uses convolution layers to detect spatial patterns efficiently compared to fully connected networks.

CNNs are specialized deep learning models designed to process structured grid data such as images. Unlike fully connected networks, CNNs exploit spatial locality using convolution filters that scan across the image. This reduces parameters, improves generalization, and captures hierarchical patterns like edges → textures → objects. Weight sharing and local receptive fields are the key principles.