Solved-We say that a random variable X belongs to a Gaussian mixture model- Solution

Description

Questions:

1. Let X₁; X₂; :::; X_n be n > 0 independent identically distributed random variables with cdf F_X (x) and pdf f_X (x) = F_X⁰(x). Derive an expression for the cdf and pdf of Y₁ = max(X₁; X₂; :::; X_n) and Y₂ = min(X₁; X₂; :::; X_n) in terms of F_X (x). [10 points]

2. We say that a random variable X belongs to a Gaussian mixture model (GMM) if X ^PK_i=1 p_iN ( _i; _i²)

where pi is the ‘mixing probability’ for each of the K constituent Gaussians, with	K	8i; 0	pi 1.
	i=1 pi = 1;

P

To draw a sample from a GMM, we do the following: (1) One of the K Gaussians is randomly chosen as per the PMF fp₁; p₂; :::; p_K g (thus, a Gaussian with a higher mixing probability has a higher chance of being picked). (2) Let the index of the chosen Gaussian be (say) m. Then, you draw the value from N ( _m; _m²). If X belongs to a GMM as de ned here, obtain expressions for E(X); Var(X) and the MGF of X.

		random variable of the form X				i N	(	; 2) for each i	1; 2; :::; K	g	. De ne another random
Now consider a		K	p	X			i	i	2 f
variable Z =	P				. Derive an expression for E(Z); Var(Z) and the PDF, MGF of Z. [2+2+2+2+2+2+3=15
points]		i=1	i	i

3. Using Markov’s inequality, prove the following one-sided version of Chebyshev’s inequality for random vari-

able X with mean and variance 2: P (X)	2		if > 0, and P (X) 1	2
		2 + 2		2 + 2

if < 0. [15 points]

4. Given stu you’ve learned in class, prove the following bounds: P (X x) e ^tx _X (t) for t > 0, and P (X x) e ^tx _X (t) for t < 0. Here _X (t) represents the MGF of random variable X for parameter t.

Now consider that X denotes the sum of n independent Bernoulli random variables X₁; X₂; :::; X_n where

[15 points]

5. Consider N independent random variables X₁; X₂; :::; X_N , such that each variable X_i takes on the values 1; 2; 3; 4; 5 with probability 0:05; 0:4; 0:15; 0:3; 0:1 respectively. For di erent values of

N 2 f5; 10; 20; 50; 100; 200; 500; 1000; 5000; 10000g, do as follows:

5. 1. Plot the (empirically determined) distribution of the average of these random variables (X_avg^(N) =

PN

_i=1 X_i=N) in the form of a histogram with 50 bins.

5. 2. Empirically determine the CDF of X_avg^(N) using the ecdf command of MATLAB (this is called the empirical CDF). On a separate gure, plot the empirical CDF. On this, overlay the CDF of a Gaussian having the same mean and variance as X_avg^(N). To get the CDF of the Gaussian, use the normcdf function of MATLAB.

5. 3. Let E^(N) denote the empirical CDF and ^(N) denote the Gaussian CDF. Compute the maximum absolute di erence (MAD) between E^(N)(x) and ^(N)(x) numerically, at all values x returned by ecdf. For this, read the documentation of ecdf carefully. Plot a graph of MAD as a function of N. [4+4+3+4 = 15 points]

5. Read in the images T1.jpg and T2.jpg from the homework folder using the MATLAB function imread and cast them as a double array. These are magnetic resonance images of a portion of the human brain, acquired with di erent settings of the MRI machine. They both represent the same anatomical structures and are perfectly aligned (i.e. any pixel at location (x; y) in both images represents the exact same physical entity). Consider random variables I₁; I₂ which denote the pixel intensities from the two images respectively. Write a piece of MATLAB code to shift the second image along the X direction by t_x pixels where t_x is an integer ranging from -10 to +10. While doing so, assign a value of 0 to unoccupied pixels. For each shift, compute the following measures of dependence between the rst image and the shifted version of the second image:

	the correlation coe cient ,
a measure of dependence called quadratic mutual information (QMI) de ned as			i1	i2 (pI1I2 (i1; i2)
	2	joint pmf) of I			1	and
	pI1 (i1)pI2 (i2)) , where pI1I2 (i1; i2) represents the normalized joint histogram (i.e.P,		P

I₂ (‘normalized’ means that the entries sum up to one).

For computing the joint histogram, use a bin-width of 10 in both I₁ and I₂. For computing the marginal histogram, you need to integrate the joint histogram along one of the two directions respectively. You should write your own joint histogram routine in MATLAB – do not use any inbuilt functions for it. Plot a graph of the values of versus t_x, and another graph of the values of QMI versus t_x.

Repeat exactly the same steps when the second image is a negative of the rst image, i.e. I₂ = 255 I₁.

Comment on all the plots. In particular, what do you observe regarding the relationship between the dependence measures and the alignment between the two images? Your report should contain all four plots labelled properly, and the comments on them as mentioned before. [30 points]

2