Hi,
I have one doubt about statements from lab file.
seems ,they are wrong.
please confirm thank you
Hi,
I have one doubt about statements from lab file.
please confirm thank you
Hi!
The explanation in the lab is correct.
As you will see in the example that the derivative of \sigma is bounded at 0.25
.
This makes the vanishing gradient issue occur when abs(largest eigenvalue) goes lower than 1/0.25 = 4
Thank you