Sampling from a discrete distribution while accounting for past occurrences - algorithm

I have a discrete distribution that I need to sample from. It is heavily skewed and an example is as follows
a - 1-40
b - 40-80
c - 80-85
d - 85-90
e - 90-95
f - 95-100
To currently sample from this distribution, I'm choosing a random number in the interval [1,100] and choosing the corresponding value.
However, I'd like to be certain that if I see one of [c,d,e,f] I don't see the exact same value being sampled for the next x samples. The context being that they are powerups in a game I'm building. I'd like the powerups to be random, but not hand out the same powerup to the player repeatedly.
Is there any method that incorporates past occurrences of samples into generating a value or do I have to repeatedly sample till I get a value I'd prefer?

One way to do this is to shuffle an array containing the values 1-100 and then iterate through as necessary.
For example, an implementation of this in Java:
public class PowerUpSelector
{
private int current;
private int x;
private int[] distribution;
public PowerUpSelector(int x)
{
this.x = x;
current = 0;
distribution = new int[100];
for (int i = 0; i < distribution.length; i++)
distribution[i] = i;
shuffle(distribution[i]); //implement fisher-yates shuffle here.
}
public int returnPowerUpID()
{
if (current >= x)
{
shuffle(distribution);
current = 0;
}
return distribution[current++];
}
}

Related

How to find trend (growth/decrease/stationarity) of a data series

I am trying to extract the OEE trend of a manufacturing machine. I already have a dataset of OEE calculated more or less every 30 seconds for each manufacturing machine and stored in a database.
What I want to do is to extract a subset of the dataset (say, last 30 minutes) and state if the OEE has grown, decreased or has been stable (withing a certain threshold). My task is NOT to forecast what will be the next value of OEE, but just to know if has decreased (desired return value: -1), grown (desired return value: +1) or been stable (desired return value: 0) based on the dataset. I am using Java 8 in my project.
Here is an example of dataset:
71.37
71.37
70.91
70.30
70.30
70.42
70.42
69.77
69.77
69.29
68.92
68.92
68.61
68.61
68.91
68.91
68.50
68.71
69.27
69.26
69.89
69.85
69.98
69.93
69.39
68.97
69.03
From this dataset is possible to state that the OEE has been decreasing (of couse based on a threshold), thus the algorithm would return -1.
I have been searching on the web unsuccessfully. I have found this, or this github project, or this stackoverflow question. However, all those are (more or less) complex forecasting algorithm. I am searching for a much easier solution. Any help is apreciated.
You could go for a
sliding average of the last n values.
Or a
sliding median of the last n values.
It highly depends on your application what is appropriate. But both these are very simple to implement and in a lot of cases more than good enough.
As you know from math, one would use d/dt, which more or less is using the step differences.
A trend is should have some weight.
class Trend {
int direction;
double probability;
}
Trend trend(double[] lastData) {
double[] deltas = Arrays.copyOf(lastData, lastData.length - 1);
for (int i = 0; i < deltas.length; ++i) {
deltas[i] -= lastData[i + 1];
}
// Trend based on two parts:
int parts = 2;
int splitN = (deltas.length + 1) / parts;
int i = 0;
int[] trends = new int[parts];
for (int j = 0; j < parts.length; ++j) {
int n = Math.min(splitN, parts.length - i);
double partAvg = DoubleStream.of(deltas).skip(i).limit(n).sum() / n;
trends[j] = tendency(partAvg);
}
Trend result = new Trend();
trend.direction = trends[parts - 1];
double avg = IntStream.of(trends).average().orElse((double)trend.direction);
trend.probability = ((direction - avg) + 1) / 2;
return trends[parts - 1];
}
int tendency(double sum) {
final double EPS = 0.0001;
return sum < -EPS ? -1 : sum > EPS ? 1 : 0;
}
This is not very sophisticated. For more elaborate treatment a math forum might be useful.

Get percentage unique numbers generated by random generator

I am using random generator in my python code. I want to get the percentage of unique random numbers generated over a huge range like from random(0:10^8).I need to generate 10^12 numbers What could be the efficient algorithm in terms of space complexity?
the code is similar to :
import random
dif = {}
for i in range(0,1000):
rannum = random.randint(0,50)
dif[rannum] = "True"
dif_len = len(dif)
print dif_len
per = float(dif_len)/50
print per
You have to keep track of each number the generator generates or there is no way to know whether some new number has been seen before. What is the best way to do that? It depends on how many numbers you are going to examine. For small N, use a HashSet. At some large number of N it becomes more efficient to use a bitmap.
For small N...
public class Accumulator {
private int uniqueNumbers = 0;
private int totalAccumulated = 0;
private HashSet<int> set = new HashSet<int>();
public void Add(int i) {
if (!set.Contains(i)) {
set.Add(i);
uniqueNumbers++;
}
totalAccumulated++;
}
public double PercentUnique() {
return 100.0 * uniqueNumbers / totalAccumulated;
}
}

implementing shuffle functioning in a music player

This was asked in an interview
"What is the most efficient way to implement a shuffle function in a music
player to play random songs without repetition"
I suggested link-list approach i.e. use a link-list, generate a random number and remove that item/song from the list ( this way , we ensure that no song is repeated )
then I suggested bit vector approach but he wasn't satisfied at all.
so what according to you is the best approach to implement such a function?
Below are some implementations. I also had difficulties during the interview but after the interview I saw that the solution is simple.
public class MusicTrackProgram {
// O(n) in-place swapping
public static List<MusicTrack> shuffle3(List<MusicTrack> input) {
Random random = new Random();
int last = input.size() - 1;
while (last >= 0) {
int randomInt = Math.abs(random.nextInt() % input.size());
// O(1)
MusicTrack randomTrack = input.get(randomInt);
MusicTrack temp = input.get(last);
// O(1)
input.set(last, randomTrack);
input.set(randomInt, temp);
--last;
}
return input;
}
// O(n) but extra field
public static List<MusicTrack> shuffle(List<MusicTrack> input) {
List<MusicTrack> result = new ArrayList<>();
Random random = new Random();
while (result.size() != input.size()) {
int randomInt = Math.abs(random.nextInt() % input.size());
// O(1)
MusicTrack randomTrack = input.get(randomInt);
if (randomTrack.isUsed) {
continue;
}
// O(1)
result.add(randomTrack);
randomTrack.isUsed = true;
}
return result;
}
// very inefficient O(n^2)
public static List<MusicTrack> shuffle2(List<MusicTrack> input) {
List<MusicTrack> result = new ArrayList<>();
Random random = new Random();
while (result.size() != input.size()) {
int randomInt = Math.abs(random.nextInt() % input.size());
// O(1)
MusicTrack randomTrack = input.get(randomInt);
// O(1)
result.add(randomTrack);
// O(n)
input.remove(randomTrack);
}
return result;
}
public static void main(String[] args) {
List<MusicTrack> musicTracks = MusicTrackFactory.generate(1000000);
List<MusicTrack> result = shuffle3(musicTracks);
result.stream().forEach(x -> System.out.println(x.getName()));
}
}
There is no perfect answer, I guess this sort of questions is aimed to start a discussion. Most likely your interviewer was wanting to hear about Fisher–Yates shuffle (aka Knuth shuffle).
Here is brief outline from wiki:
Write down the numbers from 1 through N.
Pick a random number k between one and the number of unstruck numbers remaining (inclusive).
Counting from the low end, strike out the kth number not yet struck out, and write it down elsewhere.
Repeat from step 2 until all the numbers have been struck out.
The sequence of numbers written down in step 3 is now a random permutation of the original numbers.
You should mention its inefficiencies and benefits, how you could improve this, throw in a few lines of code and discuss what and how you would test this code.
We can use link list and a queue for implementing a song search in mp3 player
We can extend this to following functionalities:
Add a new song
Delete a song
Randomly play a song
Add a song in play queue
Suppose initially we have 6 songs stored as link list
Link list has 2 pointers : start and end
totalSongCount=6
Randomly play a song:
We will generate a random number between 1 to totalSongCount. Let this be 4
We will remove the node representing song 4 and keep it after end pointer
we will decerement the totalSongCount (totalSongCount--).
Next time random number will be generated between 1 to 5 as we have decremented the totalSongCount , we can repeat the process
To add a new song, just add it to link list and make it as head pointer(add in beginning)
increment totalSongCount (totalSongCount++)
To delete a song , first find it and delete it
Also keep a track whether it is after end pointer , if it is not just decerement the totalSongCount (totalSongCount--)
The selected song can have two option:
Either play at that moment or
Add to a playlist (Seperate queue)
I think below solution should work
class InvalidInput extends Exception{
public InvalidInput(String str){
super(str);
}
}
class SongShuffler{
String songName[];
int cooldownPeriod;
Queue<String> queue;
int lastIndex ;
Random random;
public SongShuffler(String arr[], int k) throws InvalidInput{
if(arr.length < k)
throw new InvalidInput("Arr length should be greater than k");
songName = arr;
cooldownPeriod = k;
queue = new LinkedList<String>();
lastIndex = arr.length-1;
random = new Random();
}
public String getSong(){
if(queue.size() == cooldownPeriod){
String s = queue.poll();
songName[lastIndex+1] = s;
lastIndex++;
}
int ind = random.nextInt(lastIndex);
String ans = songName[ind];
queue.add(ans);
songName[ind] = songName[lastIndex];
lastIndex--;
return ans;
}
}

Printing numbers of the form 2^i * 5^j in increasing order

How do you print numbers of form 2^i * 5^j in increasing order.
For eg:
1, 2, 4, 5, 8, 10, 16, 20
This is actually a very interesting question, especially if you don't want this to be N^2 or NlogN complexity.
What I would do is the following:
Define a data structure containing 2 values (i and j) and the result of the formula.
Define a collection (e.g. std::vector) containing this data structures
Initialize the collection with the value (0,0) (the result is 1 in this case)
Now in a loop do the following:
Look in the collection and take the instance with the smallest value
Remove it from the collection
Print this out
Create 2 new instances based on the instance you just processed
In the first instance increment i
In the second instance increment j
Add both instances to the collection (if they aren't in the collection yet)
Loop until you had enough of it
The performance can be easily tweaked by choosing the right data structure and collection.
E.g. in C++, you could use an std::map, where the key is the result of the formula, and the value is the pair (i,j). Taking the smallest value is then just taking the first instance in the map (*map.begin()).
I quickly wrote the following application to illustrate it (it works!, but contains no further comments, sorry):
#include <math.h>
#include <map>
#include <iostream>
typedef __int64 Integer;
typedef std::pair<Integer,Integer> MyPair;
typedef std::map<Integer,MyPair> MyMap;
Integer result(const MyPair &myPair)
{
return pow((double)2,(double)myPair.first) * pow((double)5,(double)myPair.second);
}
int main()
{
MyMap myMap;
MyPair firstValue(0,0);
myMap[result(firstValue)] = firstValue;
while (true)
{
auto it=myMap.begin();
if (it->first < 0) break; // overflow
MyPair myPair = it->second;
std::cout << it->first << "= 2^" << myPair.first << "*5^" << myPair.second << std::endl;
myMap.erase(it);
MyPair pair1 = myPair;
++pair1.first;
myMap[result(pair1)] = pair1;
MyPair pair2 = myPair;
++pair2.second;
myMap[result(pair2)] = pair2;
}
}
This is well suited to a functional programming style. In F#:
let min (a,b)= if(a<b)then a else b;;
type stream (current, next)=
member this.current = current
member this.next():stream = next();;
let rec merge(a:stream,b:stream)=
if(a.current<b.current) then new stream(a.current, fun()->merge(a.next(),b))
else new stream(b.current, fun()->merge(a,b.next()));;
let rec Squares(start) = new stream(start,fun()->Squares(start*2));;
let rec AllPowers(start) = new stream(start,fun()->merge(Squares(start*2),AllPowers(start*5)));;
let Results = AllPowers(1);;
Works well with Results then being a stream type with current value and a next method.
Walking through it:
I define min for completenes.
I define a stream type to have a current value and a method to return a new string, essentially head and tail of a stream of numbers.
I define the function merge, which takes the smaller of the current values of two streams and then increments that stream. It then recurses to provide the rest of the stream. Essentially, given two streams which are in order, it will produce a new stream which is in order.
I define squares to be a stream increasing in powers of 2.
AllPowers takes the start value and merges the stream resulting from all squares at this number of powers of 5. it with the stream resulting from multiplying it by 5, since these are your only two options. You effectively are left with a tree of results
The result is merging more and more streams, so you merge the following streams
1, 2, 4, 8, 16, 32...
5, 10, 20, 40, 80, 160...
25, 50, 100, 200, 400...
.
.
.
Merging all of these turns out to be fairly efficient with tail recursio and compiler optimisations etc.
These could be printed to the console like this:
let rec PrintAll(s:stream)=
if (s.current > 0) then
do System.Console.WriteLine(s.current)
PrintAll(s.next());;
PrintAll(Results);
let v = System.Console.ReadLine();
Similar things could be done in any language which allows for recursion and passing functions as values (it's only a little more complex if you can't pass functions as variables).
For an O(N) solution, you can use a list of numbers found so far and two indexes: one representing the next number to be multiplied by 2, and the other the next number to be multiplied by 5. Then in each iteration you have two candidate values to choose the smaller one from.
In Python:
numbers = [1]
next_2 = 0
next_5 = 0
for i in xrange(100):
mult_2 = numbers[next_2]*2
mult_5 = numbers[next_5]*5
if mult_2 < mult_5:
next = mult_2
next_2 += 1
else:
next = mult_5
next_5 += 1
# The comparison here is to avoid appending duplicates
if next > numbers[-1]:
numbers.append(next)
print numbers
So we have two loops, one incrementing i and second one incrementing j starting both from zero, right? (multiply symbol is confusing in the title of the question)
You can do something very straightforward:
Add all items in an array
Sort the array
Or you need an other solution with more math analysys?
EDIT: More smart solution by leveraging similarity with Merge Sort problem
If we imagine infinite set of numbers of 2^i and 5^j as two independent streams/lists this problem looks very the same as well known Merge Sort problem.
So solution steps are:
Get two numbers one from the each of streams (of 2 and of 5)
Compare
Return smallest
get next number from the stream of the previously returned smallest
and that's it! ;)
PS: Complexity of Merge Sort always is O(n*log(n))
I visualize this problem as a matrix M where M(i,j) = 2^i * 5^j. This means that both the rows and columns are increasing.
Think about drawing a line through the entries in increasing order, clearly beginning at entry (1,1). As you visit entries, the row and column increasing conditions ensure that the shape formed by those cells will always be an integer partition (in English notation). Keep track of this partition (mu = (m1, m2, m3, ...) where mi is the number of smaller entries in row i -- hence m1 >= m2 >= ...). Then the only entries that you need to compare are those entries which can be added to the partition.
Here's a crude example. Suppose you've visited all the xs (mu = (5,3,3,1)), then you need only check the #s:
x x x x x #
x x x #
x x x
x #
#
Therefore the number of checks is the number of addable cells (equivalently the number of ways to go up in Bruhat order if you're of a mind to think in terms of posets).
Given a partition mu, it's easy to determine what the addable states are. Image an infinite string of 0s following the last positive entry. Then you can increase mi by 1 if and only if m(i-1) > mi.
Back to the example, for mu = (5,3,3,1) we can increase m1 (6,3,3,1) or m2 (5,4,3,1) or m4 (5,3,3,2) or m5 (5,3,3,1,1).
The solution to the problem then finds the correct sequence of partitions (saturated chain). In pseudocode:
mu = [1,0,0,...,0];
while (/* some terminate condition or go on forever */) {
minNext = 0;
nextCell = [];
// look through all addable cells
for (int i=0; i<mu.length; ++i) {
if (i==0 or mu[i-1]>mu[i]) {
// check for new minimum value
if (minNext == 0 or 2^i * 5^(mu[i]+1) < minNext) {
nextCell = i;
minNext = 2^i * 5^(mu[i]+1)
}
}
}
// print next largest entry and update mu
print(minNext);
mu[i]++;
}
I wrote this in Maple stopping after 12 iterations:
1, 2, 4, 5, 8, 10, 16, 20, 25, 32, 40, 50
and the outputted sequence of cells added and got this:
1 2 3 5 7 10
4 6 8 11
9 12
corresponding to this matrix representation:
1, 2, 4, 8, 16, 32...
5, 10, 20, 40, 80, 160...
25, 50, 100, 200, 400...
First of all, (as others mentioned already) this question is very vague!!!
Nevertheless, I am going to give a shot based on your vague equation and the pattern as your expected result. So I am not sure the following will be true for what you are trying to do, however it may give you some idea about java collections!
import java.util.List;
import java.util.ArrayList;
import java.util.SortedSet;
import java.util.TreeSet;
public class IncreasingNumbers {
private static List<Integer> findIncreasingNumbers(int maxIteration) {
SortedSet<Integer> numbers = new TreeSet<Integer>();
SortedSet<Integer> numbers2 = new TreeSet<Integer>();
for (int i=0;i < maxIteration;i++) {
int n1 = (int)Math.pow(2, i);
numbers.add(n1);
for (int j=0;j < maxIteration;j++) {
int n2 = (int)Math.pow(5, i);
numbers.add(n2);
for (Integer n: numbers) {
int n3 = n*n1;
numbers2.add(n3);
}
}
}
numbers.addAll(numbers2);
return new ArrayList<Integer>(numbers);
}
/**
* Based on the following fuzzy question # StackOverflow
* http://stackoverflow.com/questions/7571934/printing-numbers-of-the-form-2i-5j-in-increasing-order
*
*
* Result:
* 1 2 4 5 8 10 16 20 25 32 40 64 80 100 125 128 200 256 400 625 1000 2000 10000
*/
public static void main(String[] args) {
List<Integer> numbers = findIncreasingNumbers(5);
for (Integer i: numbers) {
System.out.print(i + " ");
}
}
}
If you can do it in O(nlogn), here's a simple solution:
Get an empty min-heap
Put 1 in the heap
while (you want to continue)
Get num from heap
print num
put num*2 and num*5 in the heap
There you have it. By min-heap, I mean min-heap
As a mathematician the first thing I always think about when looking at something like this is "will logarithms help?".
In this case it might.
If our series A is increasing then the series log(A) is also increasing. Since all terms of A are of the form 2^i.5^j then all members of the series log(A) are of the form i.log(2) + j.log(5)
We can then look at the series log(A)/log(2) which is also increasing and its elements are of the form i+j.(log(5)/log(2))
If we work out the i and j that generates the full ordered list for this last series (call it B) then that i and j will also generate the series A correctly.
This is just changing the nature of the problem but hopefully to one where it becomes easier to solve. At each step you can either increase i and decrease j or vice versa.
Looking at a few of the early changes you can make (which I will possibly refer to as transforms of i,j or just transorms) gives us some clues of where we are going.
Clearly increasing i by 1 will increase B by 1. However, given that log(5)/log(2) is approx 2.3 then increasing j by 1 while decreasing i by 2 will given an increase of just 0.3 . The problem then is at each stage finding the minimum possible increase in B for changes of i and j.
To do this I just kept a record as I increased of the most efficient transforms of i and j (ie what to add and subtract from each) to get the smallest possible increase in the series. Then applied whichever one was valid (ie making sure i and j don't go negative).
Since at each stage you can either decrease i or decrease j there are effectively two classes of transforms that can be checked individually. A new transform doesn't have to have the best overall score to be included in our future checks, just better than any other in its class.
To test my thougths I wrote a sort of program in LinqPad. Key things to note are that the Dump() method just outputs the object to screen and that the syntax/structure isn't valid for a real c# file. Converting it if you want to run it should be easy though.
Hopefully anything not explicitly explained will be understandable from the code.
void Main()
{
double C = Math.Log(5)/Math.Log(2);
int i = 0;
int j = 0;
int maxi = i;
int maxj = j;
List<int> outputList = new List<int>();
List<Transform> transforms = new List<Transform>();
outputList.Add(1);
while (outputList.Count<500)
{
Transform tr;
if (i==maxi)
{
//We haven't considered i this big before. Lets see if we can find an efficient transform by getting this many i and taking away some j.
maxi++;
tr = new Transform(maxi, (int)(-(maxi-maxi%C)/C), maxi%C);
AddIfWorthwhile(transforms, tr);
}
if (j==maxj)
{
//We haven't considered j this big before. Lets see if we can find an efficient transform by getting this many j and taking away some i.
maxj++;
tr = new Transform((int)(-(maxj*C)), maxj, (maxj*C)%1);
AddIfWorthwhile(transforms, tr);
}
//We have a set of transforms. We first find ones that are valid then order them by score and take the first (smallest) one.
Transform bestTransform = transforms.Where(x=>x.I>=-i && x.J >=-j).OrderBy(x=>x.Score).First();
//Apply transform
i+=bestTransform.I;
j+=bestTransform.J;
//output the next number in out list.
int value = GetValue(i,j);
//This line just gets it to stop when it overflows. I would have expected an exception but maybe LinqPad does magic with them?
if (value<0) break;
outputList.Add(value);
}
outputList.Dump();
}
public int GetValue(int i, int j)
{
return (int)(Math.Pow(2,i)*Math.Pow(5,j));
}
public void AddIfWorthwhile(List<Transform> list, Transform tr)
{
if (list.Where(x=>(x.Score<tr.Score && x.IncreaseI == tr.IncreaseI)).Count()==0)
{
list.Add(tr);
}
}
// Define other methods and classes here
public class Transform
{
public int I;
public int J;
public double Score;
public bool IncreaseI
{
get {return I>0;}
}
public Transform(int i, int j, double score)
{
I=i;
J=j;
Score=score;
}
}
I've not bothered looking at the efficiency of this but I strongly suspect its better than some other solutions because at each stage all I need to do is check my set of transforms - working out how many of these there are compared to "n" is non-trivial. It is clearly related since the further you go the more transforms there are but the number of new transforms becomes vanishingly small at higher numbers so maybe its just O(1). This O stuff always confused me though. ;-)
One advantage over other solutions is that it allows you to calculate i,j without needing to calculate the product allowing me to work out what the sequence would be without needing to calculate the actual number itself.
For what its worth after the first 230 nunmbers (when int runs out of space) I had 9 transforms to check each time. And given its only my total that overflowed I ran if for the first million results and got to i=5191 and j=354. The number of transforms was 23. The size of this number in the list is approximately 10^1810. Runtime to get to this level was approx 5 seconds.
P.S. If you like this answer please feel free to tell your friends since I spent ages on this and a few +1s would be nice compensation. Or in fact just comment to tell me what you think. :)
I'm sure everyone one's might have got the answer by now, but just wanted to give a direction to this solution..
It's a Ctrl C + Ctrl V from
http://www.careercup.com/question?id=16378662
void print(int N)
{
int arr[N];
arr[0] = 1;
int i = 0, j = 0, k = 1;
int numJ, numI;
int num;
for(int count = 1; count < N; )
{
numI = arr[i] * 2;
numJ = arr[j] * 5;
if(numI < numJ)
{
num = numI;
i++;
}
else
{
num = numJ;
j++;
}
if(num > arr[k-1])
{
arr[k] = num;
k++;
count++;
}
}
for(int counter = 0; counter < N; counter++)
{
printf("%d ", arr[counter]);
}
}
The question as put to me was to return an infinite set of solutions. I pondered the use of trees, but felt there was a problem with figuring out when to harvest and prune the tree, given an infinite number of values for i & j. I realized that a sieve algorithm could be used. Starting from zero, determine whether each positive integer had values for i and j. This was facilitated by turning answer = (2^i)*(2^j) around and solving for i instead. That gave me i = log2 (answer/ (5^j)). Here is the code:
class Program
{
static void Main(string[] args)
{
var startTime = DateTime.Now;
int potential = 0;
do
{
if (ExistsIandJ(potential))
Console.WriteLine("{0}", potential);
potential++;
} while (potential < 100000);
Console.WriteLine("Took {0} seconds", DateTime.Now.Subtract(startTime).TotalSeconds);
}
private static bool ExistsIandJ(int potential)
{
// potential = (2^i)*(5^j)
// 1 = (2^i)*(5^j)/potential
// 1/(2^1) = (5^j)/potential or (2^i) = potential / (5^j)
// i = log2 (potential / (5^j))
for (var j = 0; Math.Pow(5,j) <= potential; j++)
{
var i = Math.Log(potential / Math.Pow(5, j), 2);
if (i == Math.Truncate(i))
return true;
}
return false;
}
}

Shuffle list, ensuring that no item remains in same position

I want to shuffle a list of unique items, but not do an entirely random shuffle. I need to be sure that no element in the shuffled list is at the same position as in the original list. Thus, if the original list is (A, B, C, D, E), this result would be OK: (C, D, B, E, A), but this one would not: (C, E, A, D, B) because "D" is still the fourth item. The list will have at most seven items. Extreme efficiency is not a consideration. I think this modification to Fisher/Yates does the trick, but I can't prove it mathematically:
function shuffle(data) {
for (var i = 0; i < data.length - 1; i++) {
var j = i + 1 + Math.floor(Math.random() * (data.length - i - 1));
var temp = data[j];
data[j] = data[i];
data[i] = temp;
}
}
You are looking for a derangement of your entries.
First of all, your algorithm works in the sense that it outputs a random derangement, ie a permutation with no fixed point. However it has a enormous flaw (which you might not mind, but is worth keeping in mind): some derangements cannot be obtained with your algorithm. In other words, it gives probability zero to some possible derangements, so the resulting distribution is definitely not uniformly random.
One possible solution, as suggested in the comments, would be to use a rejection algorithm:
pick a permutation uniformly at random
if it hax no fixed points, return it
otherwise retry
Asymptotically, the probability of obtaining a derangement is close to 1/e = 0.3679 (as seen in the wikipedia article). Which means that to obtain a derangement you will need to generate an average of e = 2.718 permutations, which is quite costly.
A better way to do that would be to reject at each step of the algorithm. In pseudocode, something like this (assuming the original array contains i at position i, ie a[i]==i):
for (i = 1 to n-1) {
do {
j = rand(i, n) // random integer from i to n inclusive
} while a[j] != i // rejection part
swap a[i] a[j]
}
The main difference from your algorithm is that we allow j to be equal to i, but only if it does not produce a fixed point. It is slightly longer to execute (due to the rejection part), and demands that you be able to check if an entry is at its original place or not, but it has the advantage that it can produce every possible derangement (uniformly, for that matter).
I am guessing non-rejection algorithms should exist, but I would believe them to be less straight-forward.
Edit:
My algorithm is actually bad: you still have a chance of ending with the last point unshuffled, and the distribution is not random at all, see the marginal distributions of a simulation:
An algorithm that produces uniformly distributed derangements can be found here, with some context on the problem, thorough explanations and analysis.
Second Edit:
Actually your algorithm is known as Sattolo's algorithm, and is known to produce all cycles with equal probability. So any derangement which is not a cycle but a product of several disjoint cycles cannot be obtained with the algorithm. For example, with four elements, the permutation that exchanges 1 and 2, and 3 and 4 is a derangement but not a cycle.
If you don't mind obtaining only cycles, then Sattolo's algorithm is the way to go, it's actually much faster than any uniform derangement algorithm, since no rejection is needed.
As #FelixCQ has mentioned, the shuffles you are looking for are called derangements. Constructing uniformly randomly distributed derangements is not a trivial problem, but some results are known in the literature. The most obvious way to construct derangements is by the rejection method: you generate uniformly randomly distributed permutations using an algorithm like Fisher-Yates and then reject permutations with fixed points. The average running time of that procedure is e*n + o(n) where e is Euler's constant 2.71828... That would probably work in your case.
The other major approach for generating derangements is to use a recursive algorithm. However, unlike Fisher-Yates, we have two branches to the algorithm: the last item in the list can be swapped with another item (i.e., part of a two-cycle), or can be part of a larger cycle. So at each step, the recursive algorithm has to branch in order to generate all possible derangements. Furthermore, the decision of whether to take one branch or the other has to be made with the correct probabilities.
Let D(n) be the number of derangements of n items. At each stage, the number of branches taking the last item to two-cycles is (n-1)D(n-2), and the number of branches taking the last item to larger cycles is (n-1)D(n-1). This gives us a recursive way of calculating the number of derangements, namely D(n)=(n-1)(D(n-2)+D(n-1)), and gives us the probability of branching to a two-cycle at any stage, namely (n-1)D(n-2)/D(n-1).
Now we can construct derangements by deciding to which type of cycle the last element belongs, swapping the last element to one of the n-1 other positions, and repeating. It can be complicated to keep track of all the branching, however, so in 2008 some researchers developed a streamlined algorithm using those ideas. You can see a walkthrough at http://www.cs.upc.edu/~conrado/research/talks/analco08.pdf . The running time of the algorithm is proportional to 2n + O(log^2 n), a 36% improvement in speed over the rejection method.
I have implemented their algorithm in Java. Using longs works for n up to 22 or so. Using BigIntegers extends the algorithm to n=170 or so. Using BigIntegers and BigDecimals extends the algorithm to n=40000 or so (the limit depends on memory usage in the rest of the program).
package io.github.edoolittle.combinatorics;
import java.math.BigInteger;
import java.math.BigDecimal;
import java.math.MathContext;
import java.util.Random;
import java.util.HashMap;
import java.util.TreeMap;
public final class Derangements {
// cache calculated values to speed up recursive algorithm
private static HashMap<Integer,BigInteger> numberOfDerangementsMap
= new HashMap<Integer,BigInteger>();
private static int greatestNCached = -1;
// load numberOfDerangementsMap with initial values D(0)=1 and D(1)=0
static {
numberOfDerangementsMap.put(0,BigInteger.valueOf(1));
numberOfDerangementsMap.put(1,BigInteger.valueOf(0));
greatestNCached = 1;
}
private static Random rand = new Random();
// private default constructor so class isn't accidentally instantiated
private Derangements() { }
public static BigInteger numberOfDerangements(int n)
throws IllegalArgumentException {
if (numberOfDerangementsMap.containsKey(n)) {
return numberOfDerangementsMap.get(n);
} else if (n>=2) {
// pre-load the cache to avoid stack overflow (occurs near n=5000)
for (int i=greatestNCached+1; i<n; i++) numberOfDerangements(i);
greatestNCached = n-1;
// recursion for derangements: D(n) = (n-1)*(D(n-1) + D(n-2))
BigInteger Dn_1 = numberOfDerangements(n-1);
BigInteger Dn_2 = numberOfDerangements(n-2);
BigInteger Dn = (Dn_1.add(Dn_2)).multiply(BigInteger.valueOf(n-1));
numberOfDerangementsMap.put(n,Dn);
greatestNCached = n;
return Dn;
} else {
throw new IllegalArgumentException("argument must be >= 0 but was " + n);
}
}
public static int[] randomDerangement(int n)
throws IllegalArgumentException {
if (n<2)
throw new IllegalArgumentException("argument must be >= 2 but was " + n);
int[] result = new int[n];
boolean[] mark = new boolean[n];
for (int i=0; i<n; i++) {
result[i] = i;
mark[i] = false;
}
int unmarked = n;
for (int i=n-1; i>=0; i--) {
if (unmarked<2) break; // can't move anything else
if (mark[i]) continue; // can't move item at i if marked
// use the rejection method to generate random unmarked index j &lt i;
// this could be replaced by more straightforward technique
int j;
while (mark[j=rand.nextInt(i)]);
// swap two elements of the array
int temp = result[i];
result[i] = result[j];
result[j] = temp;
// mark position j as end of cycle with probability (u-1)D(u-2)/D(u)
double probability
= (new BigDecimal(numberOfDerangements(unmarked-2))).
multiply(new BigDecimal(unmarked-1)).
divide(new BigDecimal(numberOfDerangements(unmarked)),
MathContext.DECIMAL64).doubleValue();
if (rand.nextDouble() < probability) {
mark[j] = true;
unmarked--;
}
// position i now becomes out of play so we could mark it
//mark[i] = true;
// but we don't need to because loop won't touch it from now on
// however we do have to decrement unmarked
unmarked--;
}
return result;
}
// unit tests
public static void main(String[] args) {
// test derangement numbers D(i)
for (int i=0; i<100; i++) {
System.out.println("D(" + i + ") = " + numberOfDerangements(i));
}
System.out.println();
// test quantity (u-1)D_(u-2)/D_u for overflow, inaccuracy
for (int u=2; u<100; u++) {
double d = numberOfDerangements(u-2).doubleValue() * (u-1) /
numberOfDerangements(u).doubleValue();
System.out.println((u-1) + " * D(" + (u-2) + ") / D(" + u + ") = " + d);
}
System.out.println();
// test derangements for correctness, uniform distribution
int size = 5;
long reps = 10000000;
TreeMap<String,Integer> countMap = new TreeMap&ltString,Integer>();
System.out.println("Derangement\tCount");
System.out.println("-----------\t-----");
for (long rep = 0; rep < reps; rep++) {
int[] d = randomDerangement(size);
String s = "";
String sep = "";
if (size > 10) sep = " ";
for (int i=0; i<d.length; i++) {
s += d[i] + sep;
}
if (countMap.containsKey(s)) {
countMap.put(s,countMap.get(s)+1);
} else {
countMap.put(s,1);
}
}
for (String key : countMap.keySet()) {
System.out.println(key + "\t\t" + countMap.get(key));
}
System.out.println();
// large random derangement
int size1 = 1000;
System.out.println("Random derangement of " + size1 + " elements:");
int[] d1 = randomDerangement(size1);
for (int i=0; i<d1.length; i++) {
System.out.print(d1[i] + " ");
}
System.out.println();
System.out.println();
System.out.println("We start to run into memory issues around u=40000:");
{
// increase this number from 40000 to around 50000 to trigger
// out of memory-type exceptions
int u = 40003;
BigDecimal d = (new BigDecimal(numberOfDerangements(u-2))).
multiply(new BigDecimal(u-1)).
divide(new BigDecimal(numberOfDerangements(u)),MathContext.DECIMAL64);
System.out.println((u-1) + " * D(" + (u-2) + ") / D(" + u + ") = " + d);
}
}
}
In C++:
template <class T> void shuffle(std::vector<T>&arr)
{
int size = arr.size();
for (auto i = 1; i < size; i++)
{
int n = rand() % (size - i) + i;
std::swap(arr[i-1], arr[n]);
}
}

Resources