search

mbadry1 / DeepLearning.ai-Summary

This repository contains my personal notes and summaries on DeepLearning.ai specialization courses. I've enjoyed every little bit of the course hope you enjoy my notes too.
MIT License
5.17k stars 2.44k forks source link

Momentum formulation #171

Open sai-sai-sai opened 5 years ago

sai-sai-sai commented 5 years ago

Say in iteration t : Vdw = BetaVdw + (1-Beta)dw ---> The first term in this should actually be Vdw exponentially averaged till t-1 iterations. Since we are cacluating Vdw we can not use Vdw but need to use Vdw averaged till t-1th iteration.

mbadry1 commented 5 years ago

Do you mean we need to write it like this: Vdwt = BetaVdwt-1 + (1-Beta)dw

sai-sai-sai commented 5 years ago

Yes, that's what i think. What do you say? And i think Andrew Ng also states it as you had written in his video but it may be a typo or a quick mistake overlooked.