I have a matrix of 100x100 size and need to find the largest set of rows and columns that create a square having equal rows. Example:
A B C D E F C D E
a 0 1 2 3 4 5 a 2 3 4
b 2 9 7 9 8 2
c 9 0 6 8 9 7 ==>
d 8 9 2 3 4 8 d 2 3 4
e 7 2 2 3 4 5 e 2 3 4
f 0 3 6 8 7 2
Currently I am using this algorithm:
candidates = [] // element type is {rows, cols}
foreach row
foreach col
candidates[] = {[row], [col]}
do
retval = candidates.first
foreach candidates as candidate
foreach newRow > candidates.rows.max
foreach newCol > candidates.cols.max
// compare matrix cells in candidate to newRow and newCol
if (newCandidateHasEqualRows)
newCandidates[] = {candidate.rows+newRow, candidate.cols+newCol}
candidates = newCandidates
while candidates.count
return retval
Has anyone else come across a problem similar to this? And is there a better algorithm to solve it?
Here's the NP-hardness reduction I mentioned, from biclique. Given a bipartite graph, make a matrix with a row for each vertex in part A and a column for each vertex in part B. For every edge that is present, put a 0 in the corresponding matrix entry. Put a unique positive integer for each other matrix entry. For all s > 1, there is a Ks,s subgraph if and only if there is a square of size s (which necessarily is all zero).
Given a fixed set of rows, the optimal set of columns is easily determined. You could try the a priori algorithm on sets of rows, where a set of rows is considered frequent iff there exist as many columns that, together with the rows, form a valid square.
I've implemented a branch and bound solver for this problem in C++ at http://pastebin.com/J1ipWs5b. To my surprise, it actually solves randomly-generated puzzles of size up to 100x100 quite quickly: on one problem with each matrix cell chosen randomly from 0-9, an optimal 4x4 solution is found in about 750ms on my old laptop. As the range of cell entries is reduced down to just 0-1, the solution times get drastically longer -- but still, at 157s (for the one problem I tried, which had an 8x8 optimal solution), this isn't terrible. It seems to be very sensitive to the size of the optimal solution.
At any point in time, we have a partial solution consisting of a set of rows that are definitely included, and a set of rows that are definitely excluded. (The inclusion status of the remaining rows is yet to be determined.) First, we pick a remaining row to "try". We try including the row; then (if necessary; see below) we try excluding it. "Trying" here means recursively solving the corresponding subproblem. We record the set of columns that are identical across all rows that are definitely included in the solution. As rows are added to the partial solution, this set of columns can only shrink.
There are a couple of improvements beyond the standard B&B idea of pruning the search when we determine that we can't develop the current partial solution into a better (i.e. larger) complete solution than some complete solution we have already found:
A dominance rule. If there are any rows that can be added to the current partial solution without shrinking the set of identical columns at all, then we can safely add them immediately, and we never have to consider not adding them. This potentially saves a lot of branching, especially if there are many similar rows in the input.
We can reorder the remaining (not definitely included or definitely excluded) rows arbitrarily. So in particular, we can always pick as the next row to consider the row that would most shrink the set of identical columns: this (perhaps counterintuitive) strategy has the effect of eliminating bad combinations of rows near the top of the search tree, which speeds up the search a lot. It also happens to complement the dominance rule above, because it means that if there are ever two rows X and Y such that X preserves a strict subset of the identical columns that Y preserves, then X will be added to the solution first, which in turn means that whenever X is included, Y will be forced in by the dominance rule and we don't need to consider the possibility of including X but excluding Y.
Related
As I was reading about Knuth's Algorithm X to solve the exact cover problem, I thought of an edge case that I wanted some clarification on.
Here are my assumptions:
Given a matrix A, Algorithm X's "goal is to select a subset of the rows so that the digit 1 appears in each column exactly once."
If the matrix is empty, the algorithm terminates successfully and the solution is then the subset of rows logged in the partial solution up to that point.
If there is a column of 0's, the algorithm terminates unsuccessfully.
For reference: http://en.wikipedia.org/wiki/Algorithm_X
Consider the matrix A:
[[1 1 0]
[0 1 1]]
Steps I took:
Given Matrix A:
1. Choose a column, c, with the least number of 1's. I choose: column 1
2. Choose a row, r, that contains to a 1 in column c. I choose: row 1
3. Add r to the partial solution.
4. For each column j such that A(r, j) = 1:
For each row i such that A(i, j) = 1:
delete row i
delete column j
5. Matrix A is empty. Algorithm terminates successfully and solution is allegedly: {row 1}.
However, this is clearly not the case as row 1 only consists of [1 1 0] and clearly does not cover the 3rd column.
I would assume that the algorithm should at some point reduce the matrix to the point where there is only a single 0 and terminate unsuccessfully.
Could someone please explain this?
I think the confusion here is simply in the use of the term empty matrix. If you read Knuth's original paper (linked on the Wikipedia article you cited), you can see that he was treating the rows and columns as doubly-linked lists. When he says that the matrix is empty, he doesn't mean that it has no entries, he means that all the row and column objects have been deleted.
To clarify, I'll label the rows with lower case letters and the columns with upper case letters, as follows:
| A | B | C
---------------
a | 1 | 1 | 0
---------------
b | 0 | 1 | 1
The algorithm states that you choose a column deterministically (using any rule you wish), and he suggests choosing a column with the fewest number of 1's. So, we'll proceed as you suggest and choose column A. The only row with a 1 in column A is row a, so we choose row a and add it to the possible solution { a }. Now, row a has 1s in columns A and B, so we must delete those columns, and any rows containing 1s in those columns, that is, rows a and b, just as you did. The resulting matrix has a single column C and no rows:
| C
-------
This is not an empty matrix (it has a column remaining). However, column C has no 1s in it, so we terminate unsuccessfully, as the algorithm indicates.
This may seem odd, but it is a very important case if we intend to use an incidence matrix for the Exact Cover Problem, because columns represent elements of the set X that we wish to cover and rows represents subsets of X. So a matrix with some columns and no rows represents the exact covering problem where the collection of subsets to choose from is empty (but there are still points to cover).
If this description causes problems for your implementation, there is a simple workaround: just include the empty set in every problem. The empty set (containing no points of X) is represented by a row of all zeros. It is never selected by your algorithm as part of a solution, never collides with any other selected rows, but always ensures that the matrix is nonempty (there is at least one row) until all the columns have been deleted, which is really all you care about since you need to make sure that each column is covered by some row.
I asked whether this problem was NP-complete on the Computer Science forum, but asking for programming heuristics seems better suited for this site. So here it goes.
You are given an NxN grid of unit squares and 2N binary strings of length N. The goal is to fill the grid with 0's and 1's so that each string appears once and only once in the grid, either horizontally (left to right) or vertically (top down). Or determine that no such solution exists. If N is not fixed I suspect this is an NP-complete problem. However are there any heuristics that can hopefully speed up the search to faster than brute force trying all ways to fill in the grid with N vertical strings?
I remember programming this for my friend that had the 5x5 physical version of this game, but I used brute force back then. I can only think of this heuristic:
Consider a 4x4 map with these 8 strings (read each from left to right):
1 1 0 1
1 0 0 1
1 0 1 1
1 0 1 0
1 1 1 1
1 0 0 0
0 0 1 1
1 1 1 0
(Note that this is already solved, since the second 4 is the first 4 transposed)
First attempt:
We will choose columns from left to right. Since 7 of 8 strings start with 1, we will try to put the one with most 1s to the first column (so that we can lay rows more easily when columns are done).
In the second column, most string have 0, so you can also try putting a string with most zeros to the second row, and so on.
This i would call a wide-1 prediction, since it only looks at one column at a time
(Possible) Improvement:
You can look at 2 columns at a time (a wide-2 prediction, if i may call it like that). In this case, from the 8 strings, the most common combination of first two bits is 10 (5/8), so you would like to choose first two columns so the the combination 10 occurring as much as possible (in this case, 1111 followed by 1000 has 3 of 4 10 at start).
(Of course you don't have to stop at 2)
Weaknesses:
I don't know if this would work. I just made it up and thought it might work.
If you choose to he wide-X prediction, the number of possibilities is exponential with X
This can absolutely fail if the distribution of combinations if even.
What you can do:
As i said, this game has physical 5x5 adaptation, only there you can also lay the string from right-to-left and bottom-to-top, if you found that name, you could google further. I unfortunately don't remember it.
Sounds like you want the crossword grid filling algorithm:
First, build 2N subsets of your 2N strings -- each subset has all the strings with a particular bit at a particular postion. So subset(0,3) is all the strings that have a 0 in the 3rd position and subset(1,5) is all the strings that have a 1 in the 5th position.
The algorithm is a basic brute-force depth fist search trying all possible mappings of strings to slots in the grid, with severe pruning of impossible branches
Your search state is a set of assignments of strings to slots and a set of sets of possible assignments to the remaining slots. The initial state has 0 assignments and 2N sets, all of which contain all 2N strings.
At each step of the search, pick the most constrained set (the set with the fewest elements) from the set of possible sets. Try each element of the set in turn in that slot (adding it to the assigments and removing it from the set of sets), and constrain all the remaining sets of sets by removing the chosen string and intersecting the crossing sets with subset(X,N) (computed in step 1) where X is the bit from the chosen string and N is the row/column number of the chosen string
If you find an empty set when picking above, there is no solution with the choices so far, so backtrack up the tree to a different choice
This is still EXPTIME, but it is about as fast as you can get it. Since the main time consuming step is the set intersections, using 2N bit binary strings for your set representation is very fast -- for N=32, the sets fit in a 64-bit word and can be intersected with a single AND instruction. It also helps to have a POPCOUNT instruction, since you also need set sizes.
This can be solved as a 0/1 integer linear program with O(N^2) variables and constraints. First there are variables Xij which are 1 if string i is assigned to line j (where j=1 to N are rows and j = (N+1) to 2N are columns). Then there is a variable for each square in the grid, which indicates if the entry is 0 or 1. If the position of the square is (i,j) with variable Yij then the sum of all X variables for line j that correspond to strings that have a 1 in position i is equal to Yij, and the sum of all X variables for line j that correspond to strings that have a 0 in position i is equal to (1 - Yij). And similarly for line i and position j. Finally, the sum of all X variables Xij for each string i (summed over all lines j) is equal to 1.
There has been a lot of research in speeding up solvers for 0/1 integer programming so this may be able to often handle fairly large N (like N=100) for many examples. Also, in some cases, solving the relaxed non-integer linear program and rounding the solution off to 0/1 may produce a valid solution, in polynomial time.
We could choose the first lg 2N rows out of the 2N strings, and then since 2^(lg 2N) = 2N, in a lot of cases there shouldn't be very many ways to assign the N columns so that the prefixes of length lg 2N are respected. Then all the rows are filled in so they can be checked to see if a solution has been found. We can also try assigning more rows in the beginning, and fill in different combinations of rows besides the initial rows. (e.g. we can try filling in contiguous rows starting anywhere in the grid).
Running time for assigning lg 2N rows out of 2N strings is O((2N)^(lg 2N)) = O(2^((lg 2N)^2)), which grows slower than 2^N. Assigning columns to match the prefixes is the part that's the hardest to predict run time. If a prefix occurs K times among the assigned rows, and there are M remaining strings that have the prefix, then the number of assignments for this prefix is M*(M-1)...(M-K+1). The total number of possible column assignments is the product of these terms over all prefixes that occur among the rows. If this gets to be too large, the number of rows initially assigned can be increased. But it's hard to predict the worst-case run time unless an assumption is made like the NxN grid is filled in randomly.
I'm looking for a variation on the Hungarian algorithm (I think) that will pair N people to themselves, excluding self-pairs and reverse-pairs, where N is even.
E.g. given N0 - N6 and a matrix C of costs for each pair, how can I obtain the set of 3 lowest-cost pairs?
C = [ [ - 5 6 1 9 4 ]
[ 5 - 4 8 6 2 ]
[ 6 4 - 3 7 6 ]
[ 1 8 3 - 8 9 ]
[ 9 6 7 8 - 5 ]
[ 4 2 6 9 5 - ] ]
In this example, the resulting pairs would be:
N0, N3
N1, N4
N2, N5
Having typed this out I'm now wondering if I can just increase the cost values in the "bottom half" of the matrix... or even better, remove them.
Is there a variation of Hungarian that works on a non-square matrix?
Or, is there another algorithm that solves this variation of the problem?
Increasing the values of the bottom half can result in an incorrect solution. You can see this as the corner coordinates (in your example coordinates 0,1 and 5,6) of the upper half will always be considered to be in the minimum X pairs, where X is the size of the matrix.
My Solution for finding the minimum X pairs
Take the standard Hungarian algorithm
You can set the diagonal to a value greater than the sum of the elements in the unaltered matrix — this step may allow you to speed up your implementation, depending on how your implementation handles nulls.
1) The first step of the standard algorithm is to go through each row, and then each column, reducing each row and column individually such that the minimum of each row and column is zero. This is unchanged.
The general principle of this solution, is to mirror every subsequent step of the original algorithm around the diagonal.
2) The next step of the algorithm is to select rows and columns so that every zero is included within the selection, using the minimum number of rows and columns.
My alteration to the algorithm means that when selecting a row/column, also select the column/row mirrored around that diagonal, but count it as one row or column selection for all purposes, including counting the diagonal (which will be the intersection of these mirrored row/column selection pairs) as only being selected once.
3) The next step is to check if you have the right solution — which in the standard algorithm means checking if the number of rows and columns selected is equal to the size of the matrix — in your example if six rows and columns have been selected.
For this variation however, when calculating when to end the algorithm treat each row/column mirrored pair of selections as a single row or column selection. If you have the right solution then end the algorithm here.
4) If the number of rows and columns is less than the size of the matrix, then find the smallest unselected element, and call it k. Subtract k from all uncovered elements, and add it to all elements that are covered twice (again, counting the mirrored row/column selection as a single selection).
My alteration of the algorithm means that when altering values, you will alter their mirrored values identically (this should happen naturally as a result of the mirrored selection process).
Then go back to step 2 and repeat steps 2-4 until step 3 indicates the algorithm is finished.
This will result in pairs of mirrored answers (which are the coordinates — to get the value of these coordinates refer back to the original matrix) — you can safely delete half of each pair arbitrarily.
To alter this algorithm to find the minimum R pairs, where R is less than the size of the matrix, reduce the stopping point in step 3 to R. This alteration is essential to answering your question.
As #Niklas B, stated you are solving Weighted perfect matching problem
take a look at this
here is part of document describing Primal-dual algorithm for weighted perfect matching
Please read all and let me know if is useful to you
I have been sitting on this for almost a week now. Here is the question in a PDF format.
I could only think of one idea so far but it failed. The idea was to recursively create all connected subgraphs which works in O(num_of_connected_subgraphs), but that is way too slow.
I would really appreciate someone giving my a direction. I'm inclined to think that the only way is dynamic programming but I can't seem to figure out how to do it.
OK, here is a conceptual description for the algorithm that I came up with:
Form an array of the (x,y) board map from -7 to 7 in both dimensions and place the opponents pieces on it.
Starting with the first row (lowest Y value, -N):
enumerate all possible combinations of the 2nd player's pieces on the row, eliminating only those that conflict with the opponents pieces.
for each combination on this row:
--group connected pieces into separate networks and number these
networks starting with 1, ascending
--encode the row as a vector using:
= 0 for any unoccupied or opponent position
= (1-8) for the network group that that piece/position is in.
--give each such grouping a COUNT of 1, and add it to a dictionary/hashset using the encoded vector as its key
Now, for each succeeding row, in ascending order {y=y+1}:
For every entry in the previous row's dictionary:
--If the entry has exactly 1 group, add it's COUNT to TOTAL
--enumerate all possible combinations of the 2nd player's pieces
on the current row, eliminating only those that conflict with the
opponents pieces. (change:) you should skip the initial combination
(where all entries are zero) for this step, as the step above actually
covers it. For each such combination on the current row:
+ produce a grouping vector as described above
+ compare the current row's group-vector to the previous row's
group-vector from the dictionary:
++ if there are any group-*numbers* from the previous row's
vector that are not adjacent to any gorups in the current
row's vector, *for at least one value of X*, then skip
to the next combination.
++ any groups for the current row that are adjacent to any
groups of the previous row, acquire the lowest such group
number
++ any groups for the current row that are not adjacent to
any groups of the previous row, are assigned an unused
group number
+ Re-Normalize the group-number assignments for the current-row's
combination (**) and encode the vector, giving it a COUNT equal
to the previous row-vector's COUNT
+ Add the current-row's vector to the dictionary for the current
Row, using its encoded vector as the key. If it already exists,
then add it's COUNT to the COUNT for the pre-exising entry
Finally, for every entry in the dictionary for the last row:
If the entry has exactly one group, then add it's COUNT to TOTAL
**: Re-Normalizing simply means to re-assign the group numbers so as to eliminate any permutations in the grouping pattern. Specifically, this means that new group numbers should be assigned in increasing order, from left-to-right, starting from one. So for example, if your grouping vector looked like this after grouping ot to the previous row:
2 0 5 5 0 3 0 5 0 7 ...
it should be re-mapped to this normal form:
1 0 2 2 0 3 0 2 0 4 ...
Note that as in this example, after the first row, the groupings can be discontiguous. This relationship must be preserved, so the two groups of "5"s are re-mapped to the same number ("2") in the re-normalization.
OK, a couple of notes:
A. I think that this approach is correct , but I I am really not certain, so it will definitely need some vetting, etc.
B. Although it is long, it's still pretty sketchy. Each individual step is non-trivial in itself.
C. Although there are plenty of individual optimization opportunities, the overall algorithm is still pretty complicated. It is a lot better than brute-force, but even so, my back-of-the-napkin estimate is still around (2.5 to 10)*10^11 operations for N=7.
So it's probably tractable, but still a long way off from doing 74 cases in 3 seconds. I haven't read all of the detail for Peter de Revaz's answer, but his idea of rotating the "diamond" might be workable for my algorithm. Although it would increase the complexity of the inner loop, it may drop the size of the dictionaries (and thus, the number of grouping-vectors to compare against) by as much as a 100x, though it's really hard to tell without actually trying it.
Note also that there isn't any dynamic programming here. I couldn't come up with an easy way to leverage it, so that might still be an avenue for improvement.
OK, I enumerated all possible valid grouping-vectors to get a better estimate of (C) above, which lowered it to O(3.5*10^9) for N=7. That's much better, but still about an order of magnitude over what you probably need to finish 74 tests in 3 seconds. That does depend on the tests though, if most of them are smaller than N=7, it might be able to make it.
Here is a rough sketch of an approach for this problem.
First note that the lattice points need |x|+|y| < N, which results in a diamond shape going from coordinates 0,6 to 6,0 i.e. with 7 points on each side.
If you imagine rotating this diamond by 45 degrees, you will end up with a 7*7 square lattice which may be easier to think about. (Although note that there are also intermediate 6 high columns.)
For example, for N=3 the original lattice points are:
..A..
.BCD.
EFGHI
.JKL.
..M..
Which rotate to
A D I
C H
B G L
F K
E J M
On the (possibly rotated) lattice I would attempt to solve by dynamic programming the problem of counting the number of ways of placing armies in the first x columns such that the last column is a certain string (plus a boolean flag to say whether some points have been placed yet).
The string contains a digit for each lattice point.
0 represents an empty location
1 represents an isolated point
2 represents the first of a new connected group
3 represents an intermediate in a connected group
4 represents the last in an connected group
During the algorithm the strings can represent shapes containing multiple connected groups, but we reject any transformations that leave an orphaned connected group.
When you have placed all columns you need to only count strings which have at most one connected group.
For example, the string for the first 5 columns of the shape below is:
....+ = 2
..+++ = 3
..+.. = 0
..+.+ = 1
..+.. = 0
..+++ = 3
..+++ = 4
The middle + is currently unconnected, but may become connected by a later column so still needs to be tracked. (In this diagram I am also assuming a up/down/left/right 4-connectivity. The rotated lattice should really use a diagonal connectivity but I find that a bit harder to visualise and I am not entirely sure it is still a valid approach with this connectivity.)
I appreciate that this answer is not complete (and could do with lots more pictures/explanation), but perhaps it will prompt someone else to provide a more complete solution.
I have this homework assignment that I think I managed to solve, but am not entirely sure as I cannot prove my solution. I would like comments on what I did, its correctness and whether or not there's a better solution.
The problem is as follows: we have N groups of people, where group ihas g[i]people in it. We want to put these people on two rows of S seats each, such that: each group can only be put on a single row, in a contiguous sequence, OR if the group has an even number of members, we can split them in two and put them on two rows, but with the condition that they must form a rectangle (so they must have the same indices on both rows). What is the minimum number of seats S needed so that nobody is standing up?
Example: groups are 4 11. Minimum S is 11. We put all 4 in one row, and the 11 on the second row. Another: groups are 6 2. We split the 6 on two rows, and also the two. Minimum is therefore 4 seats.
This is what I'm thinking:
Calculate T = (sum of all groups + 1) / 2. Store the group numbers in an array, but split all the even values x in two values of x / 2 each. So 4 5 becomes 2 2 5. Now run subset sum on this vector, and find the minimum value higher than or equal to T that can be formed. That value is the minimum number of seats per row needed.
Example: 4 11 => 2 2 11 => T = (15 + 1) / 2 = 8. Minimum we can form from 2 2 11 that's >= 8 is 11, so that's the answer.
This seems to work, at least I can't find any counter example. I don't have a proof though. Intuitively, it seems to always be possible to arrange the people under the required conditions with the number of seats supplied by this algorithm.
Any hints are appreciated.
I think your solution is correct. The minimum number of seats per row in an optimal distribution would be your T (which is mathematically obvious).
Splitting even numbers is also correct, since they have two possible arrangements; by logically putting all the "rectangular" groups of people on one end of the seat rows you can also guarantee that they will always form a proper rectangle, so that this condition is met as well.
So the question boils down to finding a sum equal or as close as possible to T (e.g. partition problem).
Minor nit: I'm not sure if the proposed solution above works in the edge case where each group has 0 members, because your numerator in T = SUM ALL + 1 / 2 is always positive, so there will never be a subset sum that is greater than or equal to T.
To get around this, maybe a modulus operation might work here. We know that we need at least n seats in a row if n is the maximal odd term, so maybe the equation should have a max(n * (n % 2)) term in it. It will come out to max(odd) or 0. Since the maximal odd term is always added to S, I think this is safe (stated boldly without proof...).
Then we want to know if we should split the even terms or not. Here's where the subset sum approach might work, but with T simply equal to SUM ALL / 2.