{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

tuto08_csc5150 - Latent variables a graphical...

Info iconThis preview shows pages 1–14. Sign up to view the full content.

View Full Document Right Arrow Icon
CSC5150 Learning Theory and Computational Finance Tutorial 08 By Tu Shikui Nov.06, 2007 Some slides are from Shi Lei 2006 tutorials.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Outline FiniteMixtureModels Gaussian Mixture Model (GMM) Mean Square Error (MSE) clustering analysis Expectation Maximization (EM) algorithm How EM works? An intuitive explanation
Background image of page 2
Finite Mixture Models
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Examples for Finite Mixture Model
Background image of page 4
EM for GMM
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Latent variables -- label for the data complete incomplete Complete and Incomplete Data
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 8
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 10
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 12
Background image of page 13

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 14
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Latent variables -- a graphical representation EM in general How EM works? (1/)-- A decomposition q(z) is an arbitrary distribution of latent variable z Geometric representation of the decomposition How EM works? (2/) E-Step : max ln p ( x | θ ) min KL ( q || p ) Let q(z) = p(z|x,θ) q ( z ) M-Step : max L ( q , θ ) θ Intuitively … E-Step Intuitively … M-Step End of This Tutorial! Thank you!...
View Full Document

{[ snackBarMessage ]}