Gradient Descent Explained

Рет қаралды 81,542

IBM Technology

Күн бұрын

Пікірлер: 33

@Msyoutube38 2 жыл бұрын

Very nice explanation of the concept, brief and understandable. Awesome!

@vt1454 2 жыл бұрын

As always, great video from IBM

@John-wx3zn 9 ай бұрын

It is wrong.

@handsanitizer2457 Жыл бұрын

Wow best explanation ever 👏

@krishnakeshav23 Жыл бұрын

Good explanation. It is somewhat also important to note that curve should be differentiable.

@Akanniafelumo 2 ай бұрын

The best explanation I have had ever, in fact till now

@davidrempel433 Жыл бұрын

The most confusing part of this video is how he managed to write everything backwards on the glass so flawlessly

@sanataeeb969 Жыл бұрын

can't they write on their normal side then flip the video?

@sirpsychosexy Жыл бұрын

@@sanataeeb969 no that would be way too easy

@waliyudin86 Жыл бұрын

Bro just focus on the gradient descent topic

@P4INKiller Жыл бұрын

@@sanataeeb969Oh shit, you're clever.

@smritibasnet9782 4 ай бұрын

Nope he isnt writing backward..you can observe he seems to be using left hand to write ,but in actual right hand was being used

@krissatish87 9 ай бұрын

The best video i could find. Thank you.

@57-tycm-ii-karanshardul28 24 күн бұрын

Thankyou sir.

@cyrcesarkore 2 ай бұрын

Very simple and clear explanation. Thank you!

@Adnanuni 2 ай бұрын

Thank you for such an amazing explaination Martin. Thanks a lot team IBM

@nurudeenmohammediyam9221 3 күн бұрын

whats the difference between entropy and cost function

@Shrimant-ub4ul 6 ай бұрын

Thank You Martin , really helpful for my uni exam

@hugaexpl0it Жыл бұрын

Very good explanation of high-level concept on GD.

@harshsonar9346 Жыл бұрын

Im always confused by these screens or boards, whatever. Like how do you write on them? Do you have to write backwards or do you write normally and it kinda mirrors it?

@FaberLSH 5 ай бұрын

Thank you so much!

@sotirismoschos775 Жыл бұрын

didn't know Steve Kerr works at IBM

@s.m.rakibhasan5525 Жыл бұрын

great lecture

@SAZlearn_AI 3 ай бұрын

Let me clarify the concept of learning rate and step size in gradient descent: Learning rate: The learning rate is a hyperparameter that we set before starting the optimization process. It's a fixed value that determines how large our steps will be in general. Step size: The actual size of each step is determined by both the learning rate and the gradient at that point. Specifically: step_size = learning_rate * magnitude_of_gradient So: The learning rate itself is not the size of the steps from point to point. The learning rate is a constant that helps determine how big those steps will be. The actual size of each step can vary, even with a constant learning rate, because it also depends on the gradient at each point. To visualize this: In steep areas of the loss function (large gradient), the steps will be larger. In flatter areas (small gradient), the steps will be smaller. The learning rate acts as a general "scaling factor" for all these steps.