Видео

👍

I will start my machine learning journey, I have the knowledge of stat , calculus, linear algebra.Should I follow this series?. From comments it seems this is a very very good lecture series. Or this would be too advance for a newbie like me?

At 39:53 the example seems a little bit off to me. I would expect Newton's method to converge in a single step (the function looks like a quadratic).

👍

👍

compression comparison for cross entropy is just damn good....

Thanks a lot. Wished I could have attended 'ML for Data Science' as well

Thank you Professor !!!

"If you can't read my handwriting, you are not alone" :)

Thank you Sir !!!

Thanks a lot !!

Thank you !!

Thanks a lot !!

Thanks a lot Sir !!

Thanks !!!

Thanks !!!

Thank you Professor !!!

The time watching these lectures passes in no time !! Thanks, Professor!

Thank you for the amazing Lecture !!!

Thank you so much Sir !!!

He is a fantastic teacher. And all the energetic teaching has given him a bad throat.

I like your way of teaching more than Andrew or anyone else, cause they assume students or anyone watching is at least tiny intelligent, which just isnt the case. After all we are all just stupid, except for few people, or that one kid in class

And here i am rewatching this for the 3-th time, cause i didnt take notes. This time i wont make that mistake, i already filled a whole notebook with this. Thanks Kilian, sad there isnt anything more tho

JULIA

Super cool! Thanks!

best lecture on naive bayes.

in lecture notes there is a spelling error in the concluding summary "week classifier"

lectures so good, kilian on my poster wall. :)

Are the slides for the first part of the class available?

I have been watching your lectures for years now. I must say, the style of teaching is the best ! Every-time I need a refresher on some topic, your ML series is the first I think of. Thank you for the amazing content! 😃

wonderful lecture

A Great Beginning..excited for the next

I always tried to help, each time i did the cult took advantage now its their problem, and i dont know if i can help anybody everything has changed since last night

Raise your hand if that makes sense… crickets… ok moving on!

dead mouse got me

Prof, record small videos on SVM, SVR and Soft margin SVM, if possible.

Pure gold. The lectures are excellent. The humor is spot on. Many many thanks.

wow, you are a good teacher!

Legendary course!

I'm curious if someone was actually stealing all the notes

geez does anyone ever wonder what beast of a middle school Killian went to? 37:29

End score tallly: Under-grad = 255 Grad = 310 Grad Won!!

I cannot remember the last time I laughed so much while learning about neural networks. They usually make me cry! 😂 On the serious note, I have read NNs so many times but could not understand why putting them together was actually improving the performance. Thanks for the visualization!

Not sure why this playlist doesn't come on top when one searches for ML on youtube. Andrew Ng might be a good researcher but he's not a very good teacher. Killian teaches in such a good manner, that I never once felt bored of felt as if I'm studying. Thanks Killian, You're a Gem.

Is there any way we can get access to the projects for this course?

Amazing!!!!!!!!!!!!!!!!!!!!!!!!

We have a quiz question in lecture notes : "How does k affect the classifier? What happens if k = n? What happens if k = 1?" I do not think it is discussed in lectures. In my opinion, k is the only hyperparameter in this algorithm. For k = n, we are taking mode of the entire dataset labels as the output for test point, where as for k =1 , it will be assigned label that of the closest nearest neighbor. I have a doubt here, as we are using distance metric, what if we have 2 points(for simplicity) that are at equal distance to test point and have different labels. What happens in that case for k = 1? Similarly, for k = n, if we have equal proportion of binary class labels, how does mode works in that case?

thank you professor .

Everyone seems to have a different version. AdaGrad doesn't always work. The sum of the dot product of the gradient gets too big UNLESS one scales it down. Also, AdaGrad works best with a line search. All variations work best with a line search.

i am wondering teenager hahaha!