The revolution of CNN

(a) Regular convolution: AlexNet/VGG

(b) Separable convolution block: Split Regular convolution into Depth wise and Point wise.

(c) Separable with linear bottleneck:Import ResNet bottleneck into Separable convolution.

(d) bottleneck with expansion layer: Invert bottleneck. (Small – Large – Small)




Enjoy Reading This Article?

Here are some more articles you might like to read next:

  • Resillience
  • Multi-Head Attention
  • Preference Alignment 101
  • Challenges in Code Generation
  • PREDICTING AND OPTIMIZING LLVM COMPILER PASS ORDER