CNN (Convolutional Neural Network)

Convolutional Neural Networks apply learnable filters (kernels) that slide across input data to detect local patterns. Early layers detect low-level features like edges, middle layers combine these into shapes and textures, and deeper layers recognize complex objects. Key components include convolutional layers, pooling layers (for downsampling), and fully connected layers for final classification. CNNs revolutionized computer vision starting with AlexNet's breakthrough on ImageNet in 2012. Landmark architectures include VGGNet, ResNet (which introduced skip connections), and EfficientNet. While Vision Transformers have matched or exceeded CNNs on many benchmarks, CNNs remain widely used due to their efficiency, especially on edge devices and in real-time applications.