Upload
kanishka-khandelwal
View
774
Download
0
Embed Size (px)
DESCRIPTION
Introduction to Randomized algorithms
Citation preview
An Introduction
3/20/2012Kanishka Khandelwal-BCSE IV , JU
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Flip a coin
An algorithm which flip coins is called a randomized algorithm.
3/20/2012Kanishka Khandelwal-BCSE IV , JU
A randomized algorithm is simpler.
Making decisions could be complicated.
Consider the minimum cut problem
Randomized algorithm?
Pick a random edge and contract.And Continue until two vertices are left
3/20/2012Kanishka Khandelwal-BCSE IV , JU
A randomized algorithm is faster.
Making good decisions could be expensive.
Consider a sorting procedure.Picking an element in the middle makes the procedure very efficient,
but it is expensive (i.e. linear time) to find such an element.
5 9 13 11 8 6 7 10
5 6 7 8 9 13 11 10
Picking a random element will do.
Avoid worst-case behavior: randomness can (probabilistically) guarantee average case behavior
Efficient approximate solutions to intractable problems
In many practical problems,we need to deal with HUGE input,and don’t even have time to read it once.But can we still do something useful?
3/20/2012Kanishka Khandelwal-BCSE IV , JU
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Input Output
Random Bits
Deterministic Computer
www.lavarnd.org(doesn’t use lava lampsanymore)
Randomized algorithms make random rather than deterministic decisions.
The main advantage is that no input can reliably produce worst-case results because the algorithm runs differently each time.
These algorithms are commonly used in situations where no exact and fast algorithm is known.
Behavior can vary even on a fixed input.
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Minimum spanning trees
A linear time randomized algorithm,
but no known linear time deterministic algorithm.
Primality testing
A randomized polynomial time algorithm,
but it takes thirty years to find a deterministic one.
Volume estimation of a convex body
A randomized polynomial time approximation algorithm,
but no known deterministic polynomial time approximation algorithm.
3/20/2012Kanishka Khandelwal-BCSE IV , JU
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Monte Carlo Las Vegas
Always gives the true answer.
Running time is random.
Running time is variable whose expectation is bounded(say by a polynomail).
E.g. Randomized QuickSort Algorithm
3/20/2012Kanishka Khandelwal-BCSE IV , JU
It may produce incorrect answer!
We are able to bound its probability.
By running it many times on independent random variables, we can make the failure probability arbitrarily small at the expense of running time.
E.g. Randomized Mincut Algorithm
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Suppose we want to find a number among n given numbers which is larger than or equal to the median.
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Suppose A1 < … < An.
We want Ai, such that i ≥ n/2.
It’s obvious that the best deterministic algorithm needs O(n) time to produce the answer.
n may be very large!
Suppose n is 100,000,000,000 !
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Choose 100 of the numbers with equal probability.
find the maximum among these numbers.
Return the maximum.
3/20/2012Kanishka Khandelwal-BCSE IV , JU
The running time of the given algorithm is O(1).
The probability of Failure is 1/(2100).
Consider that the algorithm may return a wrong answer but the probability is very smaller than the hardware failure or even an earthquake!
3/20/2012Kanishka Khandelwal-BCSE IV , JU
QUICKSORT
3/20/2012Kanishka Khandelwal-BCSE IV , JU
QuickSort is a simple and efficient approach to sorting:
Select an element m from unsorted array c and divide the array into two subarrays:
csmall - elements smaller than m and
clarge - elements larger than m.
Recursively sort the subarrays and combine them together in sorted array csorted
3/20/2012Kanishka Khandelwal-BCSE IV , JU
1. QuickSort(c)2. if c consists of a single element3. return c4. m c15. Determine the set of elements csmall smaller
than m6. Determine the set of elements clarge larger
than m7. QuickSort(csmall)8. QuickSort(clarge)9. Combine csmall, m, and clarge into a single
array, csorted10. return csorted
3/20/2012Kanishka Khandelwal-BCSE IV , JU
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Runtime is based on our selection of m:
-A good selection will split c evenly such that |csmall | = |clarge |, then the runtime is O(n log n).
-For a good selection, the recurrence relation is:T(n) = 2T(n/2) + const ·n
Time it takes to split the array into 2 parts where const is a positive constant
The time it takes to sort two smaller arrays of size n/2
3/20/2012Kanishka Khandelwal-BCSE IV , JU
However, a poor selection will split c unevenly and in the worst case, all elements will be greater or less than mso that one subarray is full and the other is empty. In this case, the runtime is O(n2).
For a poor selection, the recurrence relation is:T(n) = T(n-1) + const · n
The time it takes to sort one array containing n-1 elements
Time it takes to split the array into 2 parts where const is a positive constant
QuickSort seems like an ineffecient MergeSort
To improve QuickSort, we need to choose m to be a good ‘splitter.’
It can be proven that to achieve O(nlogn) running time, we don’t need a perfect split, just reasonably good one. In fact, if both subarrays are at least of size n/4, then running time will be O(n log n).
This implies that half of the choices of m make good splitters.
3/20/2012Kanishka Khandelwal-BCSE IV , JU
To improve QuickSort, randomly select m.
Since half of the elements will be good splitters, if we choose m at random we will get a 50% chance that mwill be a good choice.
This approach will make sure that no matter what input is received, the expected running time is small.
3/20/2012Kanishka Khandelwal-BCSE IV , JU
1. RandomizedQuickSort(c)2. if c consists of a single element3. return c4. Choose element m uniformly at random from c5. Determine the set of elements csmall smaller
than m6. Determine the set of elements clarge larger than
m7. RandomizedQuickSort(csmall)8. RandomizedQuickSort(clarge)9. Combine csmall, m, and clarge into a single
array, csorted10. return csorted
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Worst case runtime: O(m2)
Expected runtime: O(m log m).
Expected runtime is a good measure of the performance of randomized algorithms, often more informative than worst case runtimes.
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Making a random choice is fast.
An adversary is powerless; randomized algorithms have no worst case inputs.
Randomized algorithms are often simpler and faster than their deterministic counterparts.
3/20/2012Kanishka Khandelwal-BCSE IV , JU
In the worst case, a randomized algorithm may be very slow.
There is a finite probability of getting incorrect answer.
However, the probability of getting a wrong answer can be made arbitrarily small by the repeated employment of randomness.
Getting true random numbers is almost impossible.
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Assignments
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Input: a set of 2D points
Determine the closest pair (and its dist)
Input points are stored in an array
Suppose we have a strange storage data structure D :
When we give a point to D, it stores the point and outputs the closest pair of points stored in D
Our knowledge: Insertion time depends on whether the closest pair is changed or not.
If output is the same: 1 clock tick
If output is not the same: |D| clock ticks
3/20/2012Kanishka Khandelwal-BCSE IV , JU
With random insertion order,
show that the expected total number of clock ticks used by D is O(n)
3/20/2012Kanishka Khandelwal-BCSE IV , JU
Suppose you are given a directed graph with n vertices and m unit-length edges. Consider the problem of estimating the number of vertices within distance d of each vertex. Give a fully polynomial approximation scheme that solves this problem simultaneously for all vertices for any fixed d.
3/20/2012Kanishka Khandelwal-BCSE IV , JU