We have a group of objects, let's call them Players. We can traverse through this group only with random order, e.g. there is no such thing as Players[0].
Each Player has a unique ID, with ID < len(Players). Player's can be added and removed to the group. When a Player gets removed it will free his ID, and if a Player gets added it will acquire an ID.
If we want to add a new Player to Players we have to generate a new unique ID. What is the fastest way to generate such ID in O(1) space?
O(n log n) is possible with binary search. Start with a = 0 and b = n. The invariant is that there exists a free id in the interval [a, b). Repeat the following until b - a = 1: let m = a + floor((b - a) / 2), count the number of ids in [a, m) and in [m, b). If [a, m) has fewer than m - a ids, then set b = m. Otherwise, set a = m.
I think you can use a Queue to enqueue the IDs that have been free'd up. Dequeue the queue to get free IDs once you have used up the highest possible ID. This will take O(1).
int highestIndex = 0;
Adding Players
if (highestIndex < len(Players)-1){
ID = ++highestIndex();
}
else if (!queue.isEmpty()){
ID = queue.dequeue();
} else{
// max players reached
}
Removing Players
queue.enqueue(ID);
Keep a boolean array. Construct a binary tree over this array, such that the leafs are the initial values in the array, and for items i, i+1 the parent is their logical AND (this means one of them is 0). When you want to insert traverse the tree from the root down to find the first empty slot (keep going left while one child is 0). This gives the first empty slot in O(log(n)). You can get O(log(log(n)) if you take each sqrt(n) group of bits and form an AND parent.
Based on question as first posed with a fixed maximum number of Players:
1) Technically the size of Players is O(1). Build a boolen array of 1000 slots, one per player, with TRUE meaning "ID is assigned". When a player dies, set the ID for his bit to false. When a new player arrives, search the bit array for a "false" bit; assign that ID to the player and set the bit.
Time is O(1), too with a big constant.
Based on question as revised with arbitrary N players:
2) Expanding Holzer's idea: keep a small fixed size array of size k < < N as a cache of free IDs. Use it the way TMJ described. [TMJ deleted his answer: it said in effect, "keep a stack of unused IDs, pop an unused one, push newly dead ones"] If the cache is empty when a new ID is needed, apply Holzer's scheme (one could even refill the small array while executing Holzer's scheme). [Sheesh, Holzer deleted his answer too, it said "try each ID in order and search the set; if nobody has that ID, use it" (O(N^2)] If the number of players arrives at more or less a steady state, this would be pretty fast because statistically there would always be some values in the fixed size array.
You can combine TMJ's idea with Per's idea, but you can't refill the array during Per's scan, only with dead player IDs.
You could put the players in a (cyclic) linked list. Deleting a player would involve cutting it out of the chain, and inserting it into another list (the "free" list). Allocating a player would cut (a random) one out of the "free" list and insert it into the "active" list.
UPDATE:
Since the array is fixed, you can use a watermark separating the allocated from the free players:
Initially {watermark = 0}
Free: {swap [this] <--> [watermark -1] ; decrement watermark; }
Allocate: {increment watermark; yield warermark-1; }
Voila!
Your question is ill-formed. The immediate answer is:
ID(newPlayer) = 1000
(You stated no requirement that the new player ID have to be less than 1000.)
More seriously, since O(1000) == O(1), you can create an array of id_seen[1000], mark all IDs you've seen so far in it, than select one you have not seen.
To make your question interesting, you have to formulate it carefully, e.g. "there are N players with IDs < K. You can only traverse the collection in unknown order. Add a new player with ID < K, using O(1) space."
One (inefficient) answer: select random number X < K. Traverse the collection. If you see a player with ID == X, restart. If you don't, use it as the new ID.
Evaluating efficiency of this algorithm for a given N and K is left as an exercise to the reader ;-)
Related
Disclaimer: I'm not a professional programmer or mathematician and this is my first time encountering the field of optimisation problems. Now that's out of the way so let's get to the problem at hand:
I got several lists, each containing various items and number called 'mandatoryAmount':
listA (mandatoryAmountA, itemA1, itemA2, itemA2, ...)
Each item has certain values (each value is a number >= 0):
itemA1 (M, E, P, C, Al, Ac, D, Ab,S)
I have to choose a certain number of items from each list determined by 'mandatoryAmount'.
Within each list I can choose every item multiple times.
Once I have all of the items from each list, I'll add up the values of each.
For example:
totalM = listA (itemA1 (M) + itemA1 (M) + itemA3 (M)) + listB (itemB1 (M) + itemB2 (M))
The goals are:
-To have certain values (totalAl, totalAc, totalAb, totalS) reach a certain number cap while going over that cap as little as possible. Anything over that cap is wasted.
-To maximize the remaining values with different weightings each
The output should be the best possible selection of items to meet the goals stated above. I imagine the evaluation function to just add up all non-waste values times their respective weightings while subtracting all wasted stats times their respective weightings.
edit:
The total amount of items across all lists should be somewhere between 500 and 1000, the number of lists is around 10 and the mandatoryAmount for each list is between 0 and 14.
Here's some sample code that uses Python 3 and OR-Tools. Let's start by
defining the input representation and a random instance.
import collections
import random
Item = collections.namedtuple("Item", ["M", "E", "P", "C", "Al", "Ac", "D", "Ab", "S"])
List = collections.namedtuple("List", ["mandatoryAmount", "items"])
def RandomItem():
return Item(
random.random(),
random.random(),
random.random(),
random.random(),
random.random(),
random.random(),
random.random(),
random.random(),
random.random(),
)
lists = [
List(
random.randrange(5, 10), [RandomItem() for j in range(random.randrange(5, 10))]
)
for i in range(random.randrange(5, 10))
]
Time to formulate the optimization as a mixed-integer program. Let's import
the solver library and initialize the solver object.
from ortools.linear_solver import pywraplp
solver = pywraplp.Solver.CreateSolver("solver", "SCIP")
Make constraints for the totals that must reach a certain cap.
AlCap = random.random()
totalAl = solver.Constraint(AlCap, solver.infinity())
AcCap = random.random()
totalAc = solver.Constraint(AcCap, solver.infinity())
AbCap = random.random()
totalAb = solver.Constraint(AbCap, solver.infinity())
SCap = random.random()
totalS = solver.Constraint(SCap, solver.infinity())
We want to maximize the other values subject to some weighting.
MWeight = random.random()
EWeight = random.random()
PWeight = random.random()
CWeight = random.random()
DWeight = random.random()
solver.Objective().SetMaximization()
Create variables and fill in the constraints. For each list there is an
equality constraint on the number of items.
associations = []
for list_ in lists:
amount = solver.Constraint(list_.mandatoryAmount, list_.mandatoryAmount)
for item in list_.items:
x = solver.IntVar(0, solver.infinity(), "")
amount.SetCoefficient(x, 1)
totalAl.SetCoefficient(x, item.Al)
totalAc.SetCoefficient(x, item.Ac)
totalAb.SetCoefficient(x, item.Ab)
totalS.SetCoefficient(x, item.S)
solver.Objective().SetCoefficient(
x,
MWeight * item.M
+ EWeight * item.E
+ PWeight * item.P
+ CWeight * item.C
+ DWeight * item.D,
)
associations.append((item, x))
if solver.Solve() != solver.OPTIMAL:
raise RuntimeError
solution = []
for item, x in associations:
solution += [item] * round(x.solution_value())
print(solution)
I think David Eisenstat has the right idea with Integer programming, but let's see if we get some good solutions otherwise and perhaps provide some initial optimization. However, I think that we can just choose all of one item in each list may make this easier to solve that it normally would be. Basically that turns it into more of a Subset Sum problem. Especially with the cap.
There are two possibilities here:
There is no solution, no condition satisfies the requirement.
There is a solution that we need to be optimized.
We really want to try to find a solution first, if we can find one (regardless of the amount of waste), then that's nice.
So let's reframe the problem: We aim to simply minimize waste, but we also need to meet a min requirement. So let's try to get as much waste as possible in ways we need it.
I'm going to propose an algorithm you could use that should work "fairly well" and is polynomial time, though could probably have some optimizations. I'll be using K to mean mandatoryAmount as it's a bit of a customary variable in this situation. Also I'll be using N to mean the number of lists. Lastly, Z to represent the total number of items (across all lists).
Get the list of all items and sort them by the amount of each value they have (first the goal values, then the bonus values). If an item has 100A, 300C, 200B, 400D, 150E and the required are [B, D], then the sort order would look like: [400,200,300,150,100]. Repeat but for one goal value. Using the same example above we would have: [400,300,150,100] for goal: D and [200,300,150,100] for goal B. Create a boolean variable for optimization mode (we start by seeking for a solution, once we find one, we can try to optimize it). Create a counter/hash to contain the unassigned items. An item cannot be unassigned more than K times (to avoid infinite loops). This isn't strictly needed, but could work as an optimization for step 5, as it prioritize goals you actually need.
For each list, keep a counter of the number of assignable slots for each list, set each to K, as well as the number of total assignable slots, and set to K * N. This will be adjusted as needed along the way. You want to be able to quickly O(1) lookup for: a) which list an (sorted) item belongs to, b) how many available slots that item has, and c) How many times has the item been unassigned, d) Find the item is the sorted list.
General Assignment. While there are slots available (total slots), go through the list from highest to lowest order. If the list for that item is available, assign as many slots as possible to that item. Update the assignable and total slots. If result is a valid solution, record it, trip the "optimization mode flag". If slots remain unassigned, revert the previous unassignment (but do not change the assignment count).
Waste Optimization. Find the most wasteful item that can be unassigned (unassigned count < K). Unassign one slot of it. If in optimization mode, do not allow any of the goal values to go below their cap (skip if it would). Update the unassigned count for item. Goto #3, but start just after the wasteful item. If no assignment made, reassign this item until the list has no remaining assignments, but do not update the unassigned count (otherwise we might end up in an invalid state).
Goal value Optimization. Skip if current state is a valid solution. Find the value furthest from it's goal (IE: A/B/C/D/E above) that can be unassigned. Unassign one slot for that item. Update assignment count. Goto step 3, begin search at start of list (unlike Step 4), stop searching the list if you go below the value of this item (not this item itself, as others may have the same value). If no assignment made, reassign this item until the list has no remaining assignments, but do not update the unassigned count (otherwise we might end up in an invalid state).
No Assignments remain. Return current state as "best solution found".
Algorithm should end with the "best" solution that this approach can come up with. Increasing max unassignment counts may improve the solution, decreasing max assignment counts will speed up the algorithm. Algorithm will run until it has maxed out it's assignment counts.
This is a bit of a greedy algorithm, so I'm not sure it's optimal (in that it will always yield the best result) but it may give you some ideas as to how to approach it. It also feels like it should yield fairly good results, as it basically trying to bound the results. Algorithm performance is something like O(Z^2 * K), where K is the mandatoryAmount and Z is the total number of items. Each item is unassigned K items, and potentially each assignment also requires O(Z) checks before it is reassigned.
As an optimization, use a O(log N) or better delete/next operation sorted data structure to store the sorted lists. Doing so it would make it practical to delete items from the assignment lists once the unassignment count reaches K (rendering them no longer assignable) allowing for O(Z * log(Z) * K) performance instead.
Edit:
Hmmm, the above only works within a single list (IE: Item removed can only be added to it's own list, as only that list has room). To avoid this, do step 4 (remove too heavy) then step 5 (remove too light) and then goto step 3 (using step 5's rules for searching, but also disallow adding back the too heavy ones).
So basically we remove the heaviest one then the lightest one then we try to assign something that is as heavy as possible to make up for the lightest one we removed.
You are in charge of a classroom which has n seats in a single row, numbered 0 through n-1
During the day students enter and leave the classroom for the exam.
In order to minimize the cheating, your task is to efficiently seat all incoming students.
You're given 2 types of queries: add_student(student_id) -> seat index and remove_student(student_id) -> void
The rules for seating the student is the following:
The seat must be unoccupied
The closest student must be as far away as possible
Ties can be resolved by choosing the lowest-numbered seat.
My Approach is to use Hashtable considering we have to assign seat index to each student which we can do using hash function. Is this approach correct?
If hashtable is the right approach then for - 'the closest student must be as far away as possible', how should I design efficient hash function?
Is there any better way to solve this problem?
This seemed like an interesting problem to me and so here is an algorithm that I have come up with, after giving it some thought.
So as the question says, it's only a single row numbered from 1 to n.
I was thinking of a greedy approach to this problem.
So here how it goes.
Person coming first will be seated at 1.
Person coming second will be seated at n, as this will create the farthest distance between them.
Now we start keeping the subsets, so we have one subset as (1,n).
Now when the third person comes in, he gets seated at mid of first subset i.e (1+n)/2. And after this we will have two subsets - (1,n/2) and (n/2,n).
Now when the fourth person arrives he can either sit in the mid of (1,n/2) and (n/2,n).
And this flow will go on.
Now suppose when a person leaves his seat and goes out of the class, then our subset range will change and then for the next incoming person, we will calculate new set of subsets for him.
Hope this helps. For this approach, I think an array of size n will also work nicely.
Update: This solution does not work because of the order the students can leave and open gaps.
So here's the solution I came up with. I would create an initial position array. So the position of the incoming student is just position[current_number_of_students-1].
Creating the array is the tricky part.
def position(number_of_chairs)
# handle edge cases
return [] if number_of_chairs.nil? || number_of_chairs <= 0
return [0] if number_of_chairs == 1
# initialize with first chair and the last chair
#position = [0, number_of_chairs - 1]
# We want to start adding the middle of row segments
# but in the correct order.
# The first segment is going to be the entire row
#segments = [0, number_of_chairs - 1]
while !#segments.empty?
current_segment = #segments.shift
mid = (current_segment[0] + current_segment[1]) / 2
if (mid > current_segment[0] && mid < current_segment[1])
#position << mid
# add the bottom half to the queue
#segments.push([current_segment[0], mid])
# add the top half to the queue
#segments.push([mid, current_segment[1]])
end
end
#position
end
This question already has answers here:
Stable separation for two classes of elements in an array
(3 answers)
Closed 9 years ago.
Suppose I have a function f and array of elements.
The function returns A or B for any element; you could visualize the elements this way ABBAABABAA.
I need to sort the elements according to the function, so the result is: AAAAAABBBB
The number of A values doesn't have to equal the number of B values. The total number of elements can be arbitrary (not fixed). Note that you don't sort chars, you sort objects that have a single char representation.
Few more things:
the sort should take linear time - O(n),
it should be performed in place,
it should be a stable sort.
Any ideas?
Note: if the above is not possible, do you have ideas for algorithms sacrificing one of the above requirements?
If it has to be linear and in-place, you could do a semi-stable version. By semi-stable I mean that A or B could be stable, but not both. Similar to Dukeling's answer, but you move both iterators from the same side:
a = first A
b = first B
loop while next A exists
if b < a
swap a,b elements
b = next B
a = next A
else
a = next A
With the sample string ABBAABABAA, you get:
ABBAABABAA
AABBABABAA
AAABBBABAA
AAAABBBBAA
AAAAABBBBA
AAAAAABBBB
on each turn, if you make a swap you move both, if not you just move a. This will keep A stable, but B will lose its ordering. To keep B stable instead, start from the end and work your way left.
It may be possible to do it with full stability, but I don't see how.
A stable sort might not be possible with the other given constraints, so here's an unstable sort that's similar to the partition step of quick-sort.
Have 2 iterators, one starting on the left, one starting on the right.
While there's a B at the right iterator, decrement the iterator.
While there's an A at the left iterator, increment the iterator.
If the iterators haven't crossed each other, swap their elements and repeat from 2.
Lets say,
Object_Array[1...N]
Type_A objs are A1,A2,...Ai
Type_B objs are B1,B2,...Bj
i+j = N
FOR i=1 :N
if Object_Array[i] is of Type_A
obj_A_count=obj_A_count+1
else
obj_B_count=obj_B_count+1
LOOP
Fill the resultant array with obj_A and obj_B with their respective counts depending on obj_A > obj_B
The following should work in linear time for a doubly-linked list. Because up to N insertion/deletions are involved that may cause quadratic time for arrays though.
Find the location where the first B should be after "sorting". This can be done in linear time by counting As.
Start with 3 iterators: iterA starts from the beginning of the container, and iterB starts from the above location where As and Bs should meet, and iterMiddle starts one element prior to iterB.
With iterA skip over As, find the 1st B, and move the object from iterA to iterB->previous position. Now iterA points to the next element after where the moved element used to be, and the moved element is now just before iterB.
Continue with step 3 until you reach iterMiddle. After that all elements between first() and iterB-1 are As.
Now set iterA to iterB-1.
Skip over Bs with iterB. When A is found move it to just after iterA and increment iterA.
Continue step 6 until iterB reaches end().
This would work as a stable sort for any container. The algorithm includes O(N) insertion/deletion, which is linear time for containers with O(1) insertions/deletions, but, alas, O(N^2) for arrays. Applicability in you case depends on whether the container is an array rather than a list.
If your data structure is a linked list instead of an array, you should be able to meet all three of your constraints. You just skim through the list and accumulating and moving the "B"s will be trivial pointer changes. Pseudo code below:
sort(list) {
node = list.head, blast = null, bhead = null
while(node != null) {
nextnode = node.next
if(node.val == "a") {
if(blast != null){
//move the 'a' to the front of the 'B' list
bhead.prev.next = node, node.prev = bhead.prev
blast.next = node.next, node.next.prev = blast
node.next = bhead, bhead.prev = node
}
}
else if(node.val == "b") {
if(blast == null)
bhead = blast = node
else //accumulate the "b"s..
blast = node
}
3
node = nextnode
}
}
So, you can do this in an array, but the memcopies, that emulate the list swap, will make it quiet slow for large arrays.
Firstly, assuming the array of A's and B's is either generated or read-in, I wonder why not avoid this question entirely by simply applying f as the list is being accumulated into memory into two lists that would subsequently be merged.
Otherwise, we can posit an alternative solution in O(n) time and O(1) space that may be sufficient depending on Sir Bohumil's ultimate needs:
Traverse the list and sort each segment of 1,000,000 elements in-place using the permutation cycles of the segment (once this step is done, the list could technically be sorted in-place by recursively swapping the inner-blocks, e.g., ABB AAB -> AAABBB, but that may be too time-consuming without extra space). Traverse the list again and use the same constant space to store, in two interval trees, the pointers to each block of A's and B's. For example, segments of 4,
ABBAABABAA => AABB AABB AA + pointers to blocks of A's and B's
Sequential access to A's or B's would be immediately available, and random access would come from using the interval tree to locate a specific A or B. One option could be to have the intervals number the A's and B's; e.g., to find the 4th A, look for the interval containing 4.
For sorting, an array of 1,000,000 four-byte elements (3.8MB) would suffice to store the indexes, using one bit in each element for recording visited indexes during the swaps; and two temporary variables the size of the largest A or B. For a list of one billion elements, the maximum combined interval trees would number 4000 intervals. Using 128 bits per interval, we can easily store numbered intervals for the A's and B's, and we can use the unused bits as pointers to the block index (10 bits) and offset in the case of B (20 bits). 4000*16 bytes = 62.5KB. We can store an additional array with only the B blocks' offsets in 4KB. Total space under 5MB for a list of one billion elements. (Space is in fact dependent on n but because it is extremely small in relation to n, for all practical purposes, we may consider it O(1).)
Time for sorting the million-element segments would be - one pass to count and index (here we can also accumulate the intervals and B offsets) and one pass to sort. Constructing the interval tree is O(nlogn) but n here is only 4000 (0.00005 of the one-billion list count). Total time O(2n) = O(n)
This should be possible with a bit of dynamic programming.
It works a bit like counting sort, but with a key difference. Make arrays of size n for both a and b count_a[n] and count_b[n]. Fill these arrays with how many As or Bs there has been before index i.
After just one loop, we can use these arrays to look up the correct index for any element in O(1). Like this:
int final_index(char id, int pos){
if(id == 'A')
return count_a[pos];
else
return count_a[n-1] + count_b[pos];
}
Finally, to meet the total O(n) requirement, the swapping needs to be done in a smart order. One simple option is to have recursive swapping procedure that doesn't actually perform any swapping until both elements would be placed in correct final positions. EDIT: This is actually not true. Even naive swapping will have O(n) swaps. But doing this recursive strategy will give you absolute minimum required swaps.
Note that in general case this would be very bad sorting algorithm since it has memory requirement of O(n * element value range).
Here is an exercise (3-15) in the book "Algorithm Design Manual".
Design a data structure that allows one to search, insert, and delete an integer X in O(1) time (i.e. , constant time, independent of the total number of integers stored). Assume that 1 ≤ X ≤ n and that there are m + n units of space available, where m is the maximum number of integers that can be in the table at any one time. (Hint: use two arrays A[1..n] and B[1..m].) You are not allowed to initialize either A or B, as that would take O(m) or O(n) operations. This means the arrays are full of random garbage to begin with, so you must be very careful.
I am not really seeking for the answer, because I don't even understand what this exercise asks.
From the first sentence:
Design a data structure that allows one to search, insert, and delete an integer X in O(1) time
I can easily design a data structure like that. For example:
Because 1 <= X <= n, so I just have an bit vector of n slots, and let X be the index of the array, when insert, e.g., 5, then a[5] = 1; when delete, e.g., 5, then a[5] = 0; when search, e.g.,5, then I can simply return a[5], right?
I know this exercise is harder than I imagine, but what's the key point of this question?
You are basically implementing a multiset with bounded size, both in number of elements (#elements <= m), and valid range for elements (1 <= elementValue <= n).
Search: myCollection.search(x) --> return True if x inside, else False
Insert: myCollection.insert(x) --> add exactly one x to collection
Delete: myCollection.delete(x) --> remove exactly one x from collection
Consider what happens if you try to store 5 twice, e.g.
myCollection.insert(5)
myCollection.insert(5)
That is why you cannot use a bit vector. But it says "units" of space, so the elaboration of your method would be to keep a tally of each element. For example you might have [_,_,_,_,1,_,...] then [_,_,_,_,2,_,...].
Why doesn't this work however? It seems to work just fine for example if you insert 5 then delete 5... but what happens if you do .search(5) on an uninitialized array? You are specifically told you cannot initialize it, so you have no way to tell if the value you'll find in that piece of memory e.g. 24753 actually means "there are 24753 instances of 5" or if it's garbage.
NOTE: You must allow yourself O(1) initialization space, or the problem cannot be solved. (Otherwise a .search() would not be able to distinguish the random garbage in your memory from actual data, because you could always come up with random garbage which looked like actual data.) For example you might consider having a boolean which means "I have begun using my memory" which you initialize to False, and set to True the moment you start writing to your m words of memory.
If you'd like a full solution, you can hover over the grey block to reveal the one I came up with. It's only a few lines of code, but the proofs are a bit longer:
SPOILER: FULL SOLUTION
Setup:
Use N words as a dispatch table: locationOfCounts[i] is an array of size N, with values in the range location=[0,M]. This is the location where the count of i would be stored, but we can only trust this value if we can prove it is not garbage. >!
(sidenote: This is equivalent to an array of pointers, but an array of pointers exposes you being able to look up garbage, so you'd have to code that implementation with pointer-range checks.)
To find out how many is there are in the collection, you can look up the value counts[loc] from above. We use M words as the counts themselves: counts is an array of size N, with two values per element. The first value is the number this represents, and the second value is the count of that number (in the range [1,m]). For example a value of (5,2) would mean that there are 2 instances of the number 5 stored in the collection.
(M words is enough space for all the counts. Proof: We know there can never be more than M elements, therefore the worst-case is we have M counts of value=1. QED)
(We also choose to only keep track of counts >= 1, otherwise we would not have enough memory.)
Use a number called numberOfCountsStored that IS initialized to 0 but is updated whenever the number of item types changes. For example, this number would be 0 for {}, 1 for {5:[1 times]}, 1 for {5:[2 times]}, and 2 for {5:[2 times],6:[4 times]}.
1 2 3 4 5 6 7 8...
locationOfCounts[<N]: [☠, ☠, ☠, ☠, ☠, 0, 1, ☠, ...]
counts[<M]: [(5,⨯2), (6,⨯4), ☠, ☠, ☠, ☠, ☠, ☠, ☠, ☠..., ☠]
numberOfCountsStored: 2
Below we flush out the details of each operation and prove why it's correct:
Algorithm:
There are two main ideas: 1) we can never allow ourselves to read memory without verifying that is not garbage first, or if we do we must be able to prove that it was garbage, 2) we need to be able to prove in O(1) time that the piece of counter memory has been initialized, with only O(1) space. To go about this, the O(1) space we use is numberOfItemsStored. Each time we do an operation, we will go back to this number to prove that everything was correct (e.g. see ★ below). The representation invariant is that we will always store counts in counts going from left-to-right, so numberOfItemsStored will always be the maximum index of the array that is valid.
.search(e) -- Check locationsOfCounts[e]. We assume for now that the value is properly initialized and can be trusted. We proceed to check counts[loc], but first we check if counts[loc] has been initialized: it's initialized if 0<=loc<numberOfCountsStored (if not, the data is nonsensical so we return False). After checking that, we look up counts[loc] which gives us a number,count pair. If number!=e, we got here by following randomized garbage (nonsensical), so we return False (again as above)... but if indeed number==e, this proves that the count is correct (★proof: numberOfCountsStored is a witness that this particular counts[loc] is valid, and counts[loc].number is a witness that locationOfCounts[number] is valid, and thus our original lookup was not garbage.), so we would return True.
.insert(e) -- Perform the steps in .search(e). If it already exists, we only need to increment the count by 1. However if it doesn't exist, we must tack on a new entry to the right of the counts subarray. First we increment numberOfCountsStored to reflect the fact that this new count is valid: loc = numberOfCountsStored++. Then we tack on the new entry: counts[loc] = (e,⨯1). Finally we add a reference back to it in our dispatch table so we can look it up quickly locationOfCounts[e] = loc.
.delete(e) -- Perform the steps in .search(e). If it doesn't exist, throw an error. If the count is >= 2, all we need to do is decrement the count by 1. Otherwise the count is 1, and the trick here to ensure the whole numberOfCountsStored-counts[...] invariant (i.e. everything remains stored on the left part of counts) is to perform swaps. If deletion would get rid of the last element, we will have lost a counts pair, leaving a hole in our array: [countPair0, countPair1, _hole_, countPair2, countPair{numberOfItemsStored-1}, ☠, ☠, ☠..., ☠]. We swap this hole with the last countPair, decrement numberOfCountsStored to invalidate the hole, and update locationOfCounts[the_count_record_we_swapped.number] so it now points to the new location of the count record.
Here is an idea:
treat the array B[1..m] as a stack, and make a pointer p to point to the top of the stack (let p = 0 to indicate that no elements have been inserted into the data structure). Now, to insert an integer X, use the following procedure:
p++;
A[X] = p;
B[p] = X;
Searching should be pretty easy to see here (let X' be the integer you want to search for, then just check that 1 <= A[X'] <= p, and that B[A[X']] == X'). Deleting is trickier, but still constant time. The idea is to search for the element to confirm that it is there, then move something into its spot in B (a good choice is B[p]). Then update A to reflect the pointer value of the replacement element and pop off the top of the stack (e.g. set B[p] = -1 and decrement p).
It's easier to understand the question once you know the answer: an integer is in the set if A[X]<total_integers_stored && B[A[X]]==X.
The question is really asking if you can figure out how to create a data structure that is usable with a minimum of initialization.
I first saw the idea in Cameron's answer in Jon Bentley Programming Pearls.
The idea is pretty simple but it's not straightforward to see why the initial random values that may be on the uninitialized arrays does not matter. This link explains pretty well the insertion and search operations. Deletion is left as an exercise, but is answered by one of the commenters:
remove-member(i):
if not is-member(i): return
j = dense[n-1];
dense[sparse[i]] = j;
sparse[j] = sparse[i];
n = n - 1
I have a stream of events flowing through my servers. It is not feasible for me to store all of them, but I would like to periodically be able to process some of them in aggregate. So, I want to keep a subset of the stream that is a random sampling of everything I've seen, but is capped to a max size.
So, for each new item, I need an algorithm to decide if I should add it to the stored set, or if I should discard it. If I add it, and I'm already at my limit, I need an algorithm to evict one of the old items.
Obviously, this is easy as long as I'm below my limit (just save everything). But how can I maintain a good random sampling without being biased towards old items or new items once I'm past that limit?
Thanks,
This is a common interview question.
One easy way to do it is to save the nth element with probability k/n (or 1, whichever is lesser). If you need to remove an element to save the new sample, evict a random element.
This gives you a uniformly random subset of the n elements. If you don't know n, you can estimate it and get an approximately uniform subset.
This is called random sampling. Source: http://en.wikipedia.org/wiki/Reservoir_sampling
array R[k]; // result
integer i, j;
// fill the reservoir array
for each i in 1 to k do
R[i] := S[i]
done;
// replace elements with gradually decreasing probability
for each i in k+1 to length(S) do
j := random(1, i); // important: inclusive range
if j <= k then
R[j] := S[i]
fi
done
A decent explanation/proof: http://propersubset.com/2010/04/choosing-random-elements.html
While this paper isn't precisely what you're looking for, it may be a good starting point in your search.
store samples in a first in first out (FIFO) queue.
set a sampling rate of so many events between samples, or randomize this a bit - depending on your patterns of events.
save every nth event, or whenever your rate tells you to, then stick it in to the end of the queue.
pop one off the top if the size is too big.
This is assuming you dont know the total number of events that will be received and that you don't need a minimum number of elements in the subset.
arr = arr[MAX_SIZE] //Create a new array that will store the events. Assuming first index 1.
counter = 1 //Initialize a counter.
while(receiving event){
random = //Generate a random number between 1 and counter
if( counter == random ){
if( counter <= MAX_SIZE ){
arr[counter] = event
}
else{
tmpRandom = //Generate a random number between 1 and MAX_SIZE
arr[tmpRandom] = event
}
}
counter =+ 1
}
Assign a probability of recording each event and store the event in an indexable data structure. When the size of the structure gets to the threshold, remove a random element and add new elements. In Ruby, you could do this:
#storage = []
prob = 0.002
while ( message = getnextMessage) do
#storage.delete((rand() * #storage.length).floor) if #storage.length > MAX_LEN
#storage << message if (rand() < prob)
end
This addresses your max size AND your non-bias toward when the event occurred. You could also choose which element gets deleted by partitioning your stored elements into buckets and then removing an element from any bucket that has more than one element. The bucket method allows you to keep one from each hour, for example.
You should also know that sampling theory is Big Math. If you need more than a layman's idea about this you should consult a qualified mathematician in your area.