Problem with large beta

[Anbu]

Hi Sir,

@paulinpaloalto @bahadir @eruzanski @Carina @neurogeek @lucapug @javier @kampamocha

In the case of exponentially weighted average, when the beta is large, it will smooth out the gradient descent steps.

So what could be the problem arise, if the beta is too too large ? what could be the problem due to high smooth updates ?

[Rashmi]

Hi Anbu,

Do you still need help with this problem?

Thanks!