A pooling layer is a fundamental component in Convolutional Neural Networks (CNNs), typically inserted between successive convolutional layers. Its primary function is to reduce the spatial dimensions (width and height) of the feature maps, thereby decreasing the number of parameters and computational cost in the network. This reduction also helps to make the features more robust to small translations, rotations, and distortions in the input image, a property known as translational invariance. The most common types of pooling operations are max pooling and average pooling. Max pooling selects the maximum value from a specific window (e.g., 2x2 or 3x3) of the feature map, while average pooling calculates the average value within that window. By down-sampling the feature maps, pooling layers contribute to controlling overfitting, improving computational efficiency, and extracting more abstract, high-level features.