[Google_Bootcamp_Day18]

Convolutional Neural Networks 2

Why convolutions?

• parameter sharing :
  A feature detector that’s useful in one part of the image is probably useful in another part of the image
• sparsity of connections :
  In each layer, each output value depends only on a small number of inputs

Training process (same as standard nn)

LeNet-5

AlexNet

VGG-16

ResNet

Problem of deep neural network : vanishing gradient problem

Residual Block

Residual Network

Why Resnet works well?

• z[l+2] , a[l] can be same dimmension by using “same” convolution
• easy to learn the identity function to just copy a[l] to a[l+2] using despite the addition of these two layers.

[source] https://www.coursera.org/learn/convolutional-neural-networks