0% found this document useful (0 votes)
50 views

Information Theory and Computing Assignment No. 1: April 10, 2020

This document contains a series of questions and solutions related to information theory and computing. It includes calculations of entropy for various random variables, conditional entropy, and mutual information. Key results include: 1) The entropy of a random variable X with 32 uniform outcomes is 5 bits. 2) The conditional entropy H(X|Y) is 1.375 bits and the conditional entropy H(Y|X) is 1.625 bits. 3) The mutual information I(X;Y) between two random variables is 0.375 bits. 4) Properties of entropy like the chain rule and the non-negativity of conditional entropy are discussed. 5) Calculations are shown for

Uploaded by

Abhishek Kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
50 views

Information Theory and Computing Assignment No. 1: April 10, 2020

This document contains a series of questions and solutions related to information theory and computing. It includes calculations of entropy for various random variables, conditional entropy, and mutual information. Key results include: 1) The entropy of a random variable X with 32 uniform outcomes is 5 bits. 2) The conditional entropy H(X|Y) is 1.375 bits and the conditional entropy H(Y|X) is 1.625 bits. 3) The mutual information I(X;Y) between two random variables is 0.375 bits. 4) Properties of entropy like the chain rule and the non-negativity of conditional entropy are discussed. 5) Calculations are shown for

Uploaded by

Abhishek Kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 12

Information Theory and Computing

Assignment No. 1

April 10, 2020

1
Que. 1.
Total number of outcomes = 32
Since, random variable is uniformly distributed,
1
thus, probability of each outcome = 32
Entropy of random variable X,
32
X
H(X) = − p(x) log2 (x)
i=1
 
1 1
= −32 · log2
32 32
1
= − log2
32
= log2 32
= 5 bits

∴ H(X) = 5 bits

2
Que. 2.
Entropy of random variable X,
X
H(X) = − p(x) log2 (x)
x 
1 1 1 1 1 1 1 1 4 1
=− · log2 + · log2 + · log2 + · log2 + · log2
2 2 4 4 8 8 16 16 64 64
1 1 1 1 1 1 4 1
= − · log2 − · log2 − · log2 − · log2
2 2 4 4 8 8 64 64
1 1 1 1 4
= · log2 2 + · log2 4 + · log2 8 + · log2 16 + · log2 64
2 4 8 16 64
1 1 1 1 4
= ·1+ ·2+ ·3+ ·4+ ·6
2 4 8 16 64
1 1 3 1 3
= + + + +
2 2 8 4 8
= 2 bits

∴ H(X) = 2 bits

3
Que. 3.
From the given marginal distribution of X,
X
H(X) = − p(x) log2 (x)
x 
1 1 1 1 1 1 1 1
=− · log2 + · log2 + · log2 + · log2
2 2 4 4 8 8 8 8
1 1 1 1 1 1 1 1
= − · log2 − · log2 − · log2 − · log2
2 2 4 4 8 8 8 8
1 1 1 1
= · log2 2 + · log2 4 + · log2 8 + · log2 8
2 4 8 8
1 1 1 1
= ·1+ ·2+ ·3+ ·3
2 4 8 8
3
=1+2·
8
7
= bits
4

∴ H(X) = 1.75 bits

4
Que. 4.
From the given marginal distribution of X,
X
H(X) = − p(x) log2 (x)
x 
1 1 1 1 1 1 1 1
=− · log2 + · log2 + · log2 + · log2
2 2 4 4 8 8 8 8
1 1 1 1 1 1 1 1
= − · log2 − · log2 − · log2 − · log2
2 2 4 4 8 8 8 8
1 1 1 1
= · log2 2 + · log2 4 + · log2 8 + · log2 8
2 4 8 8
1 1 1 1
= ·1+ ·2+ ·3+ ·3
2 4 8 8
3
=1+2·
8
7
= bits
4

∴ H(X) = 1.75 bits

Similarly from the marginal distribution of Y,


X
H(Y ) = − p(y) log2 (y)
y
 
1 1
= −4 × · log2
4 4
1
= 4 × × log2 4
4
= log2 4
= 2 bits

∴ H(Y ) = 2 bits

5
Using values of the joint probability distribution of X and Y,
XX
H(X, Y ) = − p(xi , yj ) log2 p(xi , yj )
i j
 
1 1 1 1 1 1 1 1
= · log2 + 2 × · log2 + 6 × · log2 +4× · log2
4 4 8 8 16 16 32 32
1
= × [2 × log2 4 + 2 × log2 8 + 3 × log2 16 + log2 32]
8
1
= × [2 × 2 + 2 × 3 + 3 × 4 + 5]
8
27
= bits
8
∴ H(X, Y ) = 3.375 bits

For conditional entropy H(X|Y ),


∵ H(X, Y ) = H(Y ) + H(X|Y )
∴ H(X|Y ) = H(X, Y ) − H(Y )
27
= −2
8
27 16
= −
8 8
11
= bits
8
∴ H(X|Y ) = 1.375 bits

And for conditional entropy H(Y |X),


∵ H(X, Y ) = H(X) + H(Y |X)
∴ H(Y |X) = H(X, Y ) − H(X)
27 7
= −
8 4
27 14
= −
8 8
13
= bits
8
6
∴ H(Y |X) = 1.625 bits

For mutual information I(X; Y ),

∵ I(X; Y ) = H(X) − H(X|Y )


7 11
= −
4 8
14 11
= −
8 8
3
= bits
8

∴ I(X; Y ) = 0.375 bits

7
Ques. 6.
(a) It comes from entropy’s chain rule applied to the random variable X
and g(X ), i.e. H(X, Y ) = H(X) + H(Y |X), so H(X, g(X)) = H(X) +
H(g(X)|X).

(b) Intuitively, if g(x) depends only on X and if the value of X is known,


g(x) is completely specified and it has a deterministic value. The entropy of
a deterministic value is 0, so H(g(X)|X) = 0 and H(X) + H(g(X)|X) =
H(X).

(c) Again, this formula comes from the entropy’s chain rule, in the form:
H(X, Y ) = H(Y ) + H(X|Y ).

(d) Proving that H(g(X)) + H(X|g(X) ≥ H(g(X)) means proving that


H(X|g(X)) ≥ 0: the non-negativity is one of the property of entropy and
can be proved from its definition by noting that the logarithm of a probabil-
ity (a quantity always less than or equal to 1) is non-positive. In particular
H(X|g(X)) = 0 if the knowledge of value of g(X) allows to totally specify
the value of X; otherwise H(X|g(X)) > 0 (for example if g(X) is an injective
function).

8
Ques. 7.
Compute of marginal distributions :
 
2 1
p(x) = ,
3 3
 
1 2
p(y) = ,
3 3
(a) H(X), H(Y)
 
2 2 1 1
H(X) = − · log2 + · log2
3 3 3 3
= 0.918 bits

∴ H(X) = 0.918 bits

 
1 1 2 2
H(Y ) = − · log2 + · log2
3 3 3 3
= 0.918 bits

∴ H(Y ) = 0.918 bits

(b) H(X|Y ), H(Y |X)


1
X
H(X|Y ) = p(y = i)H(X|Y = y)
i=0
1 2
= H(X|Y = 0) + H(X|Y = 1)
3 3
1 2 1 1
= H(1, 0) + H( , )
3 3 2 2
2
=
3

9
2
∴ H(X|Y ) =
3

1
X
H(Y |X) = p(x = i)H(Y |X = x)
i=0
2 1
= H(Y |X = 0) + H(Y |X = 1)
3 3
2 1 1 1
= H( , ) + H(0, 1)
3 2 2 3
2
=
3

2
∴ H(Y |X) =
3

(c) H(X,Y)
1,1
X
H(X, Y ) = p(x, y) log2 p(x, y)
x=0,y=0
1 1
= −3 · · log2
3 3
= 1.5849625 bits

(d) H(Y) - H(Y |X)


2
H(Y ) − H(Y |X) = 0.918 −
3
= 0.25134

10
(e) I(X;Y)
 
X p(x, y)
I(X; Y ) = p(x, y) log2
x,y
p(x)p(y)
 1   1   1 
1 1 1
= log2 2 1 + log2 2 2 + log2 1 3 2
3 3
3 ·
3 3
3 ·
3 3
3 ·
3 3
= 0.25162916

∴ I(X; Y ) = 0.25162916

(f) Venn Diagrams :

11
12

You might also like