Gaussian Mixture Models for Clustering

preview_player
Показать описание
Now that we provided some background on Gaussian distributions, we can turn to a very important special case of a mixture model, and one that we're going to emphasize quite a lot in this course and in the assignment, and that's called a mixture of Gaussians.

And remember that for any one of our image categories, and for any dimension of our observed vector like the blue intensity in that image, we're going to assume a Gaussian distribution to model that random variable.

So for example, for forest images, if we just look at the blue intensity, then we might have a Gaussian distribution shown with the green curve here, which is centered about this value 0.42. And I want to mention here that we're actually assuming a Gaussian for the entire three-dimensional vector RGB. And that Gaussian can have correlation structure and it will have correlation structure between these different intensities, because the amount of RGB in an image tends not to be independent, especially within a given image class. But for the sake of illustrations and keeping all the drawings simple, we're just going to look at one dimension like this blue intensity here. But really, in your head, imagine these Gaussians in this higher dimensional space.........
Рекомендации по теме
Комментарии
Автор

Thanks a lot for sharing this. It helps me a lot to understand the concept of mixture models.

emineguven
Автор

best video i've seen on this. great visuals & explanation

stewpatterson
Автор

this is a really good explanation. Where are the rest of the videos in the series?

ahsin.shabbir
Автор

Thanks! Helped a lot! Especially the visualisations!

ogonkishi
Автор

Very well presented, really got all my concepts clear 💯

vineethm
Автор

Great explanation. Thank you for the amazing work :)

nehadureja
Автор

The video sound is pretty good, beyond my imagination

hoavu
Автор

Very well explained! Thank you very much!!

bluebox
Автор

Great explanation, thank you very much !

ZinzinsIA
Автор

thanks a lot! explanation was crystal clear!

Noah-zpfn
Автор

Excellent video! Any about Bernoulli Mixture Models?

gabrielamartinezl.
Автор

you could plot 3D Gaussians, with their contours projected on the RGP plains.

MLDawn
Автор

Hey EF - randomly found this - hope all is well! Shout out to MITLL

bwessellangan
Автор

At around 6:05, the sigma_k values, they're all the same 3x3 covariance matrices right? sigma_1 == sigma_2 == sigma_3?

riasataliistiaque
Автор

This is a great video, thanks a lot for all the details!
I was wondering, in conclusion, how would the program decides if it's a sunset, a tree or a cloud picture? I am guessing it would calculate p(xi | zi=k, µk, σk) for k = 1, 2, 3, weighted by πk, and then pic the category with the highest probability?

DrSugarrr
Автор

Great Video!! very clear explanation. Does this have a part two where it is explained how it is applied using EM algorithm?

franklinsv
Автор

Why are the given weights for the distributions, are not really showcasing the distributions on the graph. I mean i would choose π1 = 45, π2 = 35, π3 = 20

LinaNielsen-ng
Автор

Why I m seeing a Gaussian curve shape in her hair😂...btw great video thank you so much

shashankkumar
Автор

you didn't explain what those histograms are in the beginning? Are they RGB histograms? What you started explaining right after made no sense because you didn't clarify how you got these histograms

muhammadsarimmehdi
welcome to shbcf.ru