Timeline for Why is the validation accuracy lower in case of CNN?
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 28, 2021 at 4:13 | comment | added | spiridon_the_sun_rotator | @user366312 for deep neural networks - with the number of layers more then 10, say, networks become very hard to optimize and accuracy saturates at rather high loss (train and test). I was a breakthrough, when Resnets were introduced, where after each layer, without downsampling or change of number of channels, one adds the input of the stack of layers to the output. Concatenation is possible as well. This combats the problems with optimization. | |
| Sep 27, 2021 at 23:34 | comment | added | user366312 | What do you mean by residual connection? | |
| Sep 27, 2021 at 22:07 | history | bounty ended | user366312 | ||
| Sep 27, 2021 at 22:07 | vote | accept | user366312 | ||
| Sep 27, 2021 at 23:33 | |||||
| Sep 22, 2021 at 4:35 | history | edited | spiridon_the_sun_rotator | CC BY-SA 4.0 |
added 936 characters in body
|
| Sep 21, 2021 at 7:12 | history | answered | spiridon_the_sun_rotator | CC BY-SA 4.0 |