Convolutional Layer
A neural network layer that applies learnable filters (kernels) across input data to detect local patterns like edges, textures, and shapes.
How It Works
A small filter (e.g., 3x3) slides across the input, computing dot products at each position. Each filter learns to detect a specific pattern. Multiple filters create a multi-channel output, each channel representing a different detected feature.
Key Properties
Parameter sharing: Same filter applied everywhere (few parameters). Translation equivariance: A pattern detected in one location works everywhere. Local connectivity: Each output depends only on a small input region.