Course Study with CSC321
[2021.03.06]
Topic : Lecture 12.3: GooLeNet (이동재)
Notes :
Links :
- https://blog.naver.com/laonple/220686328027 (GooLeNet Lecture Note 1~5)
- https://bskyvision.com/539 (Inception V1 soft review)
- https://89douner.tistory.com/62 (concatenation in Inception module)
- https://jetsonaicar.tistory.com/16 (Global Average Pooling , explained)
- https://lv99.tistory.com/21 (1X1 Conv. layer , explained)
Next : 2021.03.13 9:00 PM KST
Special Lecture : VGG Net , Res Net (Chloe)
- https://blog.naver.com/laonple/220738560542 (VGG Net [1] ~ VGG Net [2])
- https://blog.naver.com/laonple/220761052425 (Res Net [1] ~ Res Net [3])
Will Cover
VGG Net : using only 3X3 kernel ? (what is factorizing colvolution filter ?)
VGG Net : how to deal with gradient vanish/exploding problem (pre-trained kernel initializing)
VGG Net : technique on how-to train/test dataset (scale jittering) <- 어려운 개념이니 간단하게만
Rest Net : what is residual learning? ( Shortcut-connection? Identity mapping?)
Rest Net : what features resnet team took from VGG? (common vs. diff)
Rest Net : BottleNeck Layer (only for models with layers>50)
Rest Net : other experiment with CIFAR dataset (going for 1000 layers) ***
