MoG,II2011 - 3/1/2011 Algorithm and its derivation MOG,II...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
3/1/2011 1 MOG,II Algorithm and its derivation WHEN PDF SAMPLE IS UNAVAILABLE ´ Now, we consider the case when the values of PDF f ( x ) is not available ´ But assume: we can sample a random variable X~ f ( x ) ´ It is indeed more convenient! ´ A prototype of MCMC ´ A good example for HMM ´ The EM algorithm works (to some degree)
Background image of page 1

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

View Full DocumentRight Arrow Icon
3/1/2011 2 A BABY EXAMPLE ´ Problem: Model the PDF of the height of students ´ Given: measurement results x 1 , x 2 , x 3 , …, x N ´ Assumption: Each height follows a Gaussian g ( x; μ 1 , σ 1 ) or g ( x; μ 2 , σ 2 ) depending on the sex ´ Easy case: we know the sex, say Male: x 1 , x 4 ,..., x N-1 Female: x 2 , x 3 ,…, x N Then it’s easy to estimate the means and variances of each Gaussian ´ Hard case: we do not know the sex of the person Note that we can not sample the PDF itself directly: what is the value of f ( x )? Samples of X~ f ( x ) EXAMPLE ´ Assume p(male)= p and p(female)= q ´ X~ f(x)=pg(x; μ 1 , σ 1 )+qg(x; μ 2 , σ 2 ) ´ What is p(male|x i )?
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 01/16/2012 for the course MAD 4103 taught by Professor Li during the Spring '11 term at University of Central Florida.

Page1 / 8

MoG,II2011 - 3/1/2011 Algorithm and its derivation MOG,II...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online