Robust Principal Component Analysis (RPCA)

Danke great video!

Finally people can now distinguish Clark Kent from Superman! I thought it is never gonna happen

By 0-norm do you mean the number of non-zero entries? Thanks

Hi Professor you are so handsome that I really enjoy your video like a TV drama!

I'm doing POD which is based on PCA. Is their constraint PCA?

Very informative. Great video.

Brilliant! Would this work with kernel PCA as well?

Is a 2011 pub, recent? Appreciate video but couldn't help but ask.

Wow great video

Awesome as usual

I studied PCA last week. And now this. 😆

Wonderful explanation! Loved the way you explained this with the help of NETFLIX problem

World need more people like you

"very cool, a little bit alarming, but I'm going to walk you through it." Wait, doesn't that mean he might be admitting he's irresponsible?? Good grief. What does he expect people to think?.. "well he's telling us how to do potentially really bad stuff but that's okay cuz he's also telling us it might be bad."

thanks

What even is that? Calculus? Statistics? Geometry? What do I google If I wanna learn that maths?

Could u talking about architecture robot interactive/creative and AI

Hello Mr. Steve, please, what is the features that RPCA extracted it from image?

Hello Mr. Brunton, please, in your book, "Data Driven Science & Engineering " in page 124, in RPCA Code, in "while" instruction, why you use "count < 1000" ? what is you mean by 1000 ?

Need a detailed lecture series on RPCA, you are a gem sir
Thank you for such amazing explanation

its like a ship but one person is absurdly fat

Nice video. It is amazing how RPCA introduces Robustness in front of huge differences. I have a question regarding to your choice of mu. In your code you are choosing mu as mu = n1*n2/(4*sum(abs(X(:)))); where does this expression come from?

can i ask what is the brand of black T-shirt?
I am searching for a good quality T-shirt and stick with it

kAk∗ + λkEk1 is the convex, can you explain that

i love your explanations, they are so eloquent and fluent! thank you!

I really appreciate your help!

I find it quite dangerous that cops would be able to such software to arrest someone and potentially convict them. You know as well as I do that you cannot know what is hidden. If someone has a disguise on you cannot remove that disguise and know what you are seeing is actually what they are without some type of apriori info. Take your picture of your dude. He could have a mole... and no amount of reconstruction will ever know if he has one or not. He could have scar, have another mustache, etc. It is impossible for any reconstructive algorithm EVER to reconstruct such a thing. The information simply isn't there. Law enforcement will take your claims and being ignorant of how they work, will assume it must be true. They will use these tools to arrest people falsely and ruin their life. We've already seen them do such things.

All one can do with such algorithms is reconstruct the missing data in a way that convinces us it would fit(plausible), it can't find what it "should be"(although if it is trained on all peoples faces, for example then it could more likely find a match but this has apriori info). It's one thing for it to be used as a way to find better tailored approximations to specific problems where the end result isn't crucial... it is entirely something else to pull a rabbit out of a hat and claim it is real magic.

Such algorithms would be, say, great for movies or imagine compression when lossy is ok... but they should never be used when lives are at stake.

Thanks. Then is there any reason to use regular PCA at all?

You are amazing! Your explanations are impeccable! Thank you!

Do I understand correctly that this method does not help reduce the data dimension?

How exactly is this algorithm trained? I mean, nowhere in the given calculations it was required to have several observations, one matrix was enough. Why can't we just take a picture and extract the right components from it?

How do you create "allFaces.mat " from the yale database so I can follow along in the book? I got the database, but am not sure how to easily import it to matlab.

Thank you for the great video. I am very interested in the Netflix example (sounds like a missing value imputation problem) but couldn't find any resources/papers explaining it. I am mostly interested in using RPCA for missing value imputation in time-series. Could you please share some materials on that subject?

why low rank matrix represent normal data?

Thank you so much for this video. It was very eye opening for getting into ML

Is there any available implementation in python? Kind regards.

Someone does not like The Big Lebowski?

Thanks a lot for sharing your knowledge.

I cant download or open tem PDF book. Someone are having the same problem?

I think you are doing a really really great job here!

My good lord, where have I been living? This was EXCEPTIONAL !! Thank you

Very clear explanation. Thank you! I was wondering whether you could tell a bit about the set up you are using to do these explanations. Especially the ability to present behind the transparent board is very good (does not cover the board by the body of the lecturer). I am an educator myself and this information will be very helpful for me.

Big Fan sir thank you :)