**Pixels and features:** Images are grids of numbers; vision systems learn to extract hierarchical features (edges, textures, shapes, objects) from those grids - **Convolutional neural networks (CNNs):** Architectures that learn visual patterns through layers of filters, from simple features to comp