Відео

Thanks for the course

Thank you for this video!

Very useful lecture. Thanks for sharing! It may be a small typo (at 18:25), should be l = loss(output,target) # "output" instead of "input"

I didn't understand why adding gaussian noise to image does affect the accuracy of discriminative model. (I think it doesn't) In reality models are robust to noise

Absolutely amazing! Thank you so much!!

Loved it! Clear explanations with simple examples.

Finally i have intuitive view of seld_attention . Thank you😇

This is exceptionally well explained. I'm a student at TU Delft and this really helped me understand how to speed up my code, and why it works. Thanks a lot!

Absolutely amazing series of videos! Congrats!

Very well explained

i dont understand why we have summation in conditionals , should that be a product instead of summation ?

quite bad explanation , just repeating slides text

the approach of the noise filter presented in the video can cause neuron saturation i guess because having high weighted input with maximize the output of the sigmoid function.. which is not desirable

Thanks for your sharing! Nice and clear video!

Currently I am doing research work on TCN

Superb Superb explanation, Superb explanation thanks Superb explanation thanks a Superb explanation thanks a lot! With no predictions now: Wish you all the best and am very grateful for your work.

Why do these videos have ads on them

Google should rank videos according to the likes and the number of previously viewed videos on the same topics: this should go straight to the top for Attention/Transformer searches because I have seen and read plenty, and this is the first time the QKV as dictionary vs RDBMs made sense; that confusion had been so bad it literally stopped me thinking every time I had to consider Q, or K, or V and thus prevented me grokking the big idea. I now want to watch/read everything by you.

Watching this video feels like trying to decipher alien scriptures with a blindfold on.

Waiting for more lectures..

Self attension animationua-cam.com/video/WusQB464qMY/v-deo.htmlsi=NBxi02yTPzSfMCb6

Hello can someone explain me this, the key and the values for each iteration wont it be the same, as we compare it to 5:29 , please help me on this

Best explanation ever! Congratulations and thank you!

what the dimensions of first conv filter in AlexNet - is it 11x11x 1 or it 11x11x 3 ?

I loved u explanation !!!

You are a great teacher.

Great lecture, thanks!

Really good series of mini-lectures, thanks!

Brilliant content and explanation! It's unfortunate that the fourth part of this lecture series was never published, this realization has left me very sad in 2023. :(

love this two times

Loved this, thanks a lot now I undersand better this things and feel more confident to jump into the code part.

Such a clear explanation! Thank you!

Thanks for the great explanation! Just one question, if simple self-attention has no parameters, how can we expect it to learn? it is not trainable.

Is this ASMR?

good tutorial

Better than the Karpathy explainer video. Enough said!

holy shit, been trying to wrap my head around self-attention for a while, but it all finally clicked together with this video. very well explained, very good video :)

Great video. Thanks

Hi I think there are several things in the video that I don't know if they are correct. 1. Computation graph of transformer block: In the original paper, it says that layer normalization is performed AFTER adding output of attention layer to input of attention layer, yet in your presentation 2:22, it seems that you perform layer normalization before self attention. 2. in 3:10, I'm confused whether gamma and beta are vectors or scalars. If gamma is vector, then gamma times x should give a scalar, but that seems not true...| Can you please clarify these questions? Albeit, I'm benefiting so much from your videos. Thanks for sharing!

so far the best video describing this topic. Only questoin i have is how do we get around the fact that a word will have highest self attention with it self. You said you would clarify about this but I could not find this point.

extremely underrated resource for learning backpropagation! Thank you for this!

I think one of the best videos describing self-attention. Thank you for sharing.

Great lecture! I really appreciated your presentation by starting with simple self-attention, very helpful.

best explanation i found for self attention and multi head attention on internet , thank you sir

wow

brilliant....nothing less !

Models are based on historic information. There’s no bias.

Hi extremely helpful video here, I really appreciate but i have a question i dont understand how multi head self attention works if we are not generate extra parameters for each stack of self attention layer, what is the difference in each stack so that we can grasp the different relations of the same word in each layer

Hello sir, am very thankful for your videos learnt alot from them. I just need some help in how to use TCN for anomaly detection for multivariate time series data.

The best deep learning introductory video i had watched. And I cannot understand why such a great content only has 7 likes.