CS231n课程笔记：Leture9 CNN Architecture

 The pooling layer! 0 parameters!!!why?

The parameters are the weights right that we're trying to learn 

AlexNet

 Why use smaller filters? (3*3)

Stack of three 3x3 conv (stride 1) layers has same effective receptive field as one 7x7 conv layer

What is the effective receptive field of three 3x3 conv(stride 1) layers?

It's able to have more non-linearities in there , and it's also fewer parameters.

the parameter detaile

 

 Inception module

 

      

Solution:"bottleneck" layers that use 1x1 convolutions to reduce feature depth

 

 

 ResNet!!

 gradient blow up

 

 

 setting detailes