Weighted Moving average

Course Q&A Deep Learning Specialization Improving Deep Neural Networks: Hyperparameter tun
1

I have a very basic doubt :
in the exponentially weighted averages section, why are we computing Current temperature V_t when we already have it in the form of Theta_t. Basically why are we predicting when we have the actual?

Also in general if we want to predict V_t shouldn’t we be using Beta*V_t-1 + (1-Beta)*V_t-2?

2

Hi, @samarthdhawan.

You’re not trying to make predictions, but rather smooth out the values of image.

Here’s a good explanation by @jonaslalin :slight_smile:

1 Like