Neural Networks Demystified [Part 3: Gradient Descent]

Neural Networks Demystified
@stephencwelch

Supporting Code:

Link to Yann's Talk:

In this short series, we will build and train a complete Artificial Neural Network in python. New videos every other friday.

Part 1: Data + Architecture
Part 2: Forward Propagation
Part 3: Gradient Descent
Part 4: Backpropagation
Part 5: Numerical Gradient Checking
Part 6: Training
Part 7: Overfitting, Testing, and Regularization

[sameenfatima]

bruh. this is better than ANYTHING on the internet made since. You did a great great job 9 years ago. This is literally so easy, simple and beautifully put together. The playlist needs to make a comeback and go viral.

[ahmedas]

whenever I wanna refresh my understanding of neural networks i come back here and watch those videos all over again. thanks so much.

[kelkarmhr]

By far the best explanation for neural networks I have heard so far. We need more teachers like you. :)

[KMG_Capital]

You deserve an award for the clarity, depth, and pace, all of which means, with little effort any layman could understand what neural network real is. Thank you man!!!

[im_tanmay_g]

Dude, your are simply awesome! I don't know since how long I was looking for something like this - short, simple & totally understandable. Kudos to your hard work.

[kevinmcinerney]

These are all absolutely amazing, Mr Welsh. You have the gift of clarity.Thank you.

[JoaoCavalcanti]

Dear Stephen, just started watching your videos about neural netwoks. They are just great! Please, keep doing more videos!

[lmvejst]

I'm a totally noob in Neural Network, I came here just because I'm taking a deep learning course. This video is intuitive and awesome !

[sunilp]

Omg, this is so beautifully understandable I'm going to cry. Saving my life for this neural computing exam...

[nilaymehta]

This was really lucid and simple to understand. Quite unlike the mathematically dense explanations that one is otherwise used to

[mostinho]

Done thanks
5:50 stochastic gradient descent vs batch gradient descent
Stochastic calculates the derivative of our cost function wrt weights for each input, batch calculates it for all inputs

[AidanYi]

This has been so far the most clear explanation of neural network. Thank you.

[ninjaasmoke]

Here in 2021 December, preparing for my exams watching these videos. Thanks

[ZardicharSC]

Wow, this series deserves vastly more views!

[joshbrenneman]

These videos are genius. They really helped me visualize what a neural network does behind the scenes. Thank you!!!

[ivanhernandezaguilar]

I've been in a course of machine learning and neural networks for along 2 months and i had a lot of question until i found your videos, thank you so much

[adamyatripathi]

OK, this is the single most awesome tutorial on neural network!! Amazing job +Welch Labs !!
This shows your passion of teaching and also how clear your concepts are. You are an inspiration to teachers and professors. Love your method of instruction! I sincerely request you to upload more videos for other algorithms of machine learning. Thank you so much for these!

[yichen]

Really good and easy-to-understand demonstration.

[himanshu]

Totally understood the topics better than when my professor taught ! Kudos!

[nullindullghooomin]

Beautifully Explained. We can imagine the hard work behind each video for such neat presentation.