Gaussian Mixture Models for Clustering
HTML-код
- Опубликовано: 29 сен 2024
- 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.........
Thanks a lot for sharing this. It helps me a lot to understand the concept of mixture models.
Nice, Emily!
this is a really good explanation. Where are the rest of the videos in the series?
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
Excellent video! Any about Bernoulli Mixture Models?
Thanks! Helped a lot! Especially the visualisations!
you could plot 3D Gaussians, with their contours projected on the RGP plains.
Very well presented, really got all my concepts clear 💯
Bhai mujhe iska samajh nhi Aaya
This is gold, thank you so much
At around 6:05, the sigma_k values, they're all the same 3x3 covariance matrices right? sigma_1 == sigma_2 == sigma_3?
Hey EF - randomly found this - hope all is well! Shout out to MITLL
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?
Great explanation. Thank you for the amazing work :)
very nice tutorial. Thanks a lot.
Very well explained! Thank you very much!!
Thank you so much for this
best video i've seen on this. great visuals & explanation
Great Video!! very clear explanation. Does this have a part two where it is explained how it is applied using EM algorithm?
You may search in Coursera!
thanks a lot! explanation was crystal clear!
very well explanation
The video sound is pretty good, beyond my imagination
Great explanation
Great explanation, thank you very much !
Amazing
That's great thanks
Great!!!
Great video
thx! it is very helpful.
beautiful explanation
macam mana nak buat?
great video thx!
Thank you!
This is just what I needed.
I'm a student that finds it very difficult to find materials that suit me (balance of intuition/ mathematical detail, pace etc.). But this type of teaching of yours works wonders for me. I will watch anything you're willing to teach :)
Why I m seeing a Gaussian curve shape in her hair😂...btw great video thank you so much
Very well explained!
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
She did explain. those histograms are the distributions of the blue channel of the images in the 3 clusters.