CS322
-
Upload
luke-schwartz -
Category
Documents
-
view
24 -
download
1
description
Transcript of CS322
Logical warmup
This is a puzzle we should have done with sequences
Consider the following sequence, which should be read from left to right, starting at the top row
11 12 1
1 2 1 11 1 1 2 2 1
What are the next two rows in the sequence?
Combinations example
How many ways are there to choose 5 people out of a group of 12?
What if two people don't get along? How many 5 person teams can you make from a group of 12 if those two people cannot both be on the team?
Poker examples
How many five-card poker hands contain two pairs?
If a five-card hand is dealt at random from an ordinary deck of cards, what is the probability that the hand contains two pairs?
r-combinations with repetitions
What if you want to take r things out of a set of n things, but you are allowed to have repetitions?
Think of it as putting r things in n categories
Example: n = 5, r = 4
We could represent this as x||xx|x| That's an r x's and n – 1 |'s
1 2 3 4 5
x xx x
r-combinations with repetitions
So, we can think of taking an r-combination with repetitions as choosing r items in a string that is r + n – 1 long and marking those as x's
Consequently, the number of r-combinations with repetitions is
r
nr 1
Example
Let's say you grab a handful of 10 Starbursts
Original Starbursts come in Cherry Lemon Strawberry Orange
How many different handfuls are possible?
How many possible handfuls will contain at least 3 cherry?
Handy dandy guide to counting
This is a quick reminder of all the different ways you can count things:
Order Matters Order Doesn't Matter
Repetition Allowed nk
Repetition Not Allowed P(n,k)
k
nk 1
k
n
Pascal's Triangle
Hopefully, you are all familiar with Pascal's Triangle, the beginning of which is:
If we number rows and columns starting at 0, note that the value of row n, column r is exactly
1
1 1
1 2 1
1 3 3 1
1 4 6 4 1
1 5 10 10 5 1
1 6 15 20 15 6 1
r
n
Pascal's Formula
Pascal's Triangle works because of Pascal's Formula:
We can easily show its truth:
r
n
r
n
r
n
1
1
)!1(!)!1(
)!1(!)1(!
)!1(!)1(!
)!1(!!
)!(!!
)!1()!1(!
1
rnrn
rnrnn
rnrrnn
rnrrn
rnrn
rnrn
r
n
r
n
Binomial Theorem
a + b is called a binomial Using combinations (or Pascal's
Triangle) it is easy to compute (a + b)n
We could prove this by induction, but you probably don't care
kknn
k
n bak
nba
0
)(
Probability axioms
Let A and B be events in the sample space S 0 ≤ P(A) ≤ 1 P() = 0 and P(S) = 1 If A B = , then P(A B) = P(A) + P(B) It is clear then that P(Ac) = 1 – P(A) More generally, P(A B) = P(A) + P(B) – P(A B)
All of these axioms can be derived from set theory and the definition of probability
Union probability example
What is the probability that a card drawn randomly from an Anglo-American 52 card deck is a face card (jack, queen, or king) or is red (hearts or diamonds)?
Hint: Compute the probability that it is a face
card Compute the probability that it is red Compute the probability that it is both
Expected value
Expected value is one of the most important concepts in probability, especially if you want to gamble
The expected value is simply the sum of all events, weighted by their probabilities
If you have n outcomes with real number values a1, a2, a3, … an, each of which has probability p1, p2, p3, … pn, then the expected value is:
n
kkkpa
1
Expected value: Roulette
A normal American roulette wheel has 38 numbers: 1 through 36, 0, and 00
18 numbers are red, 18 numbers are black, and 0 and 00 are green
The best strategy you can have is always betting on black (or red)
If you bet $1 on black and win, you get $1, but you lose your dollar if it lands red or green
What is the expected value of a bet?
Conditional probability
Given that some event A has happened, the probability that some event B will happen is called conditional probability
This probability is:
)()(
)|(APBAP
ABP
Conditional probability example
Given two, fair, 6-sided dice, what is the probability that the sum of the numbers they show when rolled is 8, given that both of the numbers are even?
Bayes' Theorem
Let sample space S be a union of mutually disjoint events B1, B2, B3, … Bn
Let A be an event in S Let A and B1 through Bn have non-
zero probabilities For Bk where 1 ≤ k ≤ n
)()|(...)()|()()|()()|(
)|(2211 nn
kkk BPBAPBPBAPBPBAP
BPBAPABP
Applying Bayes' theorem
Bayes' theorem is often used to evaluate tests that can have false positives and false negatives
Consider a test for a disease that 1 in 5000 people have The false positive rate is 3% The false negative rate is 1%
What's the probability that a person who tests positive for the disease has the disease?
Let A be the event that the person tests positively for the disease
Let B1 be the event that the person actually has the disease
Let B2 be the event that the person does not have the disease
Apply Bayes' theorem
Independent events
If events A and B are events in a sample space S , then these events are independent if and only ifP(A B) = P(A)∙P(B)
This should be clear from conditional probability
If A and B are independent, then P(B|A) = P(B)
)()()()(
)()()|(
BAPBPAPAPBAP
BPABP