0PricingLogin▼
Deep Learning Academy · Lesson

VGG: Stacks of Small Filters

Depth from simple 3x3 blocks.

One Filter Size to Rule

VGG made a bold bet: use only tiny 3x3 convolutions everywhere, and just stack a lot of them.

Why Small Wins

Two stacked 3x3 layers see the same area as one 5x5, but with fewer parameters and an extra nonlinearity in between.

All lessons in this course

  1. LeNet & AlexNet: The First Wins
  2. VGG: Stacks of Small Filters
  3. ResNet: Skip Connections Go Deep
  4. Load torchvision Models
← Back to Deep Learning Academy