Course Information
Lecture Timings:
My Office:
Room # A-312
Office Hours
TIME / DAYS
Monday
Tuesday
5:00pm-5:50pm
Advanced Digital
System Design
Stochastic
Systems
IAEC (CR#20)
Wednesday
Advanced Digital
System Design
6:00pm-6:50pm
7:00pm-7:50pm
8:00pm-8:50pm
Advanced Digital
Communication
Systems
IAEC Lec Hall
Library/Make-up
Class/Seminar
Advanced Digital
Signal Processing
Library/Make-up
Class/Seminar
Thursday
Friday
Advanced Digital
Communication
Systems
IAEC Lec Hall
Library/Make-up
Class/Seminar
Stochastic Systems
IAEC (CR#20)
Library/Make-up
Class/Seminar
Library/Make-up
Class/Seminar
Saturday
Library/Make-up
Class/Seminar
5:00pm-5:50pm
6:00pm-6:50pm
7:00pm-7:50pm
8:00pm-8:50pm
Monday
Tuesday
Adv. Computer
Networks
Stochastic
Systems
IAEC (CR#20)
Advanced Digital
Communication
Systems
IAEC Lec Hall
EMC/EMI
RIMMS CR#22
Library/Make-up
Class/Seminar
Library/Make-up
Class/Seminar
Wednesday
Adv. Computer
Networks
EMC/EMI
RIMMS CR#22
Library/Make-up
Class/Seminar
Thursday
Friday
Saturday
Library/Make-up
Class/Seminar
Library/Make-up
Class/Seminar
Advanced Digital
Communication
Systems
IAEC Lec Hall
Stochastic Systems
IAEC (CR#20)
Library/Make-up
Class/Seminar
Textbooks
Albert Leon-Garcia
Sheldon M. Ross
Garnett P. Williams
Course Outline
Syllabus
Stochastic Processes
Prediction and Estimation
Markov Chain
Counting processes, queueing theory (time permitting)
Grading
(subject to change)
Final Exam:
Midterm Exam:
Quizzes:
Homework Assignments:
50%
30%
10%
10%
Policies
Attendance:
Will be taken at any time during class.
The current rules of the school will be followed (min. 75%
attendance requirement).
Boilerplate Disclaimers
We will cover:
Independence
Law of Total Probability
Bayes Theorem
Definition of Probability
Probability:
1 : the quality or state of being possible
2 : something (as an event or circumstance) that is possible
3 : the ratio of the number of outcomes in an exhaustive set of
equally likely outcomes that produce a given event to the
total number of possible outcomes, the chance that a given
event will occur
We will revisit these definitions in a little bit
10
A procedure
An outcome
Procedure
(e.g., flipping a coin)
Outcome
(e.g., the value
observed [head, tail] after
flipping the coin)
Sample Space
(Set of All Possible
Outcomes)
11
12
s1
s5
s6
s4
s2
s3
13
Experiment:
Sample space:
= {1, 2, 3, 4, 5, 6}
Events:
14
Axioms of Probability
Mutual Exclusivity
s5
s1
s6
A2
s2
s4
s3
S
Find Pr{1 2} and Pr{1} + Pr{2} in the fair dice
example
Copyright Syed Ali Khayam 2008
16
Mutual Exclusivity
Pr{1 2} =? ?
s5
s1
s6
s4
s2
s3
S
Copyright Syed Ali Khayam 2008
17
Mutual Exclusivity
s1
s6
s4
s2
s3
S
Copyright Syed Ali Khayam 2008
18
Conditional Probability
s5
s1
Event B has
already occurred
s6
s2
s4
s3
s5
s1
s6
s2
s4
s3
19
Conditional Probability
s5
s1
s6
s2
s4
Event B has
already
occurred
s3
s5
s1
s6
s4
s2
s3
= {1 , 5 , 6 , }
| = = {6 }
20
Conditional Probability
Pr
s5
s1
s4
s3
s5
s1
s6
s2
Pr
Pr{|} =
Pr{}
Pr{6 }
1/6
1
=
=
=
Pr{}
3/6
3
Event has
already
occurred
s6
s4
s2
s3
= {1, 5, 6}
Copyright Syed Ali Khayam 2008
21
Independence
In other words, the fact that B has already happened does not
affect the probability of As outcomes
Implications:
(|) = ()
= ()
()
( ) = () ()
22
Independence: Example
s2
s6
s4
s5
s1
s3
S
Copyright Syed Ali Khayam 2008
23
Independence: Example
Are events and independent?
Check if
( ) = () ()
s2
s6
s4
s5
s1
s3
S
Copyright Syed Ali Khayam 2008
24
Independence: Example
Yes!
s2
s6
s4
s5
s1
s3
S
Copyright Syed Ali Khayam 2008
25
Independence: Example
s2
s6
s4
s5
s1
s3
S
Copyright Syed Ali Khayam 2008
26
Independence: Example
No!
s2
s6
s4
s5
s1
s3
S
Copyright Syed Ali Khayam 2008
27
28
29
S={
A1
30
31
32
33
34
35
36
37
Recap
1.
2.
3.
In general, we have:
38
Recap
4.
5.
6.
39
T0
T1
R0
Pr{R1|T1}
R1
40
Pr{R0|T0}
T0
T1
R0
Pr{R1|T1}
R1
41
T1
R0
Pr{R1|T1}
R1
42
T0
T1
R0
Pr{R1|T1}
R1
1 01
Pr = Pr Pr = 01 =
2
2
1 1 1
Pr Pr = =
2 2 4
Copyright Syed Ali Khayam 2008
43
Total Probability
= 1 2
= ,
B2
B1
s1
B3
B4
s2
s5
s4
s6
s3
44
Total Probability
B2
s1
B1
s5
A
B4
s2
s4
s6
B3
s3
45
Total Probability
1 2 = and 1 2 =
B2
s1
B1
s5
A
B4
s2
s4
s6
B3
s3
46
Total Probability
1 2 . . = and 1 2 =
B2
s1
B1
s5
A
B4
s2
s4
s6
B3
s3
47
Total Probability
1 2 . . = and 1 2 =
= ( 1) ( 2) ( )
B2
s1
B1
s5
A
B4
s2
s4
s6
B3
s3
48
Total Probability
1 2 . . = and 1 2 =
= ( 1) ( 2) ( )
B2
s1
B1
s5
A
B4
s2
s4
s6
B3
s3
49
Total Probability
1 2 . . = and 1 2 =
= ( 1) ( 2) ( )
B2
s1
B1
s5
A
B4
s2
s4
s6
B3
s3
50
Total Probability
B2
s1
B1
s5
A
B4
s2
s4
s6
B3
s3
51
B2
s1
B1
s5
A
B4
s2
s4
s6
B3
s3
52
Bayes Theorem
B2
s1
B1
s5
A
B4
s2
s4
s6
B3
s3
53
Bayes Theorem
s1
B1
s5
A
B4
s2
s4
s6
B3
s3
54
Bayes Theorem
Pr Pr
Pr =
Pr{}
From the Law of Total Probability, we have:
Pr{} = Pr{|1} Pr{1} + Pr{|2} Pr{2} + + Pr{|} Pr{}
Bayes Rule
B2
s1
B1
s5
A
B4
s2
s4
s6
B3
s3
55
Bayes Theorem
s5
A
B4
s2
s4
s6
B3
s3
56
Bayes Theorem
Pr | Pr
Pr =
=
=1 Pr | Pr
Pr{ |}
Pr{| }
likelihood probability
Pr{ }
Pr{}
57
A Fishy Problem
Asad is a fisherman.
Asad is an educated person.
Asad builds a fishing robot that will do his work for him.
B2
B3
B1
B4
58
A Fishy Problem
Question: If Asads robot catches a fish that is detected red, what
species is it, 1 , 2 , 3 or 4 ?
Answer: It could be any of four species in the sea.
B2
B3
B1
B4
59
A Fishy Problem
Lets change the question:
What is the probability that a red fish is a species B1, B2,
B3 and B4?
B2
B3
B1
B4
60
A Fishy Problem
B2
B3
B1
B4
61