Solutions to Matrix System with Integer Constraints

  • Context: Graduate 
  • Thread starter Thread starter srodniki
  • Start date Start date
Click For Summary
SUMMARY

The discussion focuses on determining the number of solutions for a system of equations represented by a matrix with integer constraints. Specifically, it addresses the equation format a_{ij} where the sum of each row equals k_{i} and each column equals s_{j}. The initial solution count without column restrictions is given by the product of binomial coefficients, specifically ∏_{i=1}^{m} binom{n+k_{i}-1}{k_{i}}. The challenge arises when both row and column constraints are applied, leading to a decrease in the number of valid solutions. The discussion suggests that combinatorial techniques may be necessary to generalize the solution count.

PREREQUISITES
  • Understanding of combinatorial mathematics, specifically binomial coefficients.
  • Familiarity with integer programming and matrix equations.
  • Knowledge of constraints in mathematical systems.
  • Experience with generating functions and their applications in combinatorics.
NEXT STEPS
  • Explore advanced combinatorial techniques for constrained systems.
  • Study integer programming methods for solving matrix equations.
  • Learn about generating functions and their role in counting solutions.
  • Investigate the application of the Inclusion-Exclusion Principle in combinatorial problems.
USEFUL FOR

Mathematicians, combinatorial theorists, and data scientists interested in solving complex systems of equations with integer constraints.

srodniki
Messages
2
Reaction score
0
Hi there,
I know that the number of solutions of [itex]a_{1}+a_{2}+...a_{n}=k[/itex] where [itex]a_{i}[/itex] are non-negative integers is just [itex]\binom{n+k-1}{k}[/itex].

Now, given a [itex]m\times n[/itex] matrix with non negative integers [itex]a_{ij}[/itex], how many solutions are there for the following kind of system?
[itex]\begin{array}{ccccccccc}<br /> s_{1} & & s_{2} & & & & s_{n}\\<br /> \shortparallel & & \shortparallel & & & & \shortparallel\\<br /> a_{11} & + & a_{12} & + & ... & + & a_{1n} & = & k_{1}\\<br /> + & & + & & & & +\\<br /> a_{21} & + & a_{22} & + & ... & + & a_{2n} & = & k_{2}\\<br /> + & & + & & & & +\\<br /> \vdots & & \vdots & & & & \vdots\\<br /> + & & + & & & & +\\<br /> a_{m1} & + & a_{m2} & + & ... & + & a_{mn} & = & k_{m}<br /> \end{array}[/itex]

The sum of each row and each column is constrained by the known non-negative integers [itex]k_{i}[/itex] and [itex]s_{i}[/itex].

Without the column restriction we would have [itex]\prod_{i=1}^{m}\binom{n+k_{i}-1}{k_{i}}[/itex].

Thanks a lot.
 
Physics news on Phys.org
I would say that the size of the solution decreases the more constraints you put on

the values. If my generic a_mn has to satisfy two equations ; horizontal and vertical,

then it becomes harder to find such a value.
 
Last edited:
Shouldn't there be a combinatorial trick? I mean here is an example for [itex]\left(i_{1},i_{2}\right)=\left(5,3\right)[/itex] and [itex]n=3[/itex] . The system becomes

[itex]\begin{array}{ccccccc}<br /> s_{1} & & s_{2} & & s_{3}\\<br /> \shortparallel & & \shortparallel & & \shortparallel\\<br /> a_{11} & + & a_{12} & + & a_{13} & = & 5\\<br /> + & & + & & +\\<br /> a_{21} & + & a_{22} & + & a_{23} & = & 3<br /> \end{array}[/itex]

Starting considering only the horizontal equations, trivially the solutions are any combination of the following two columns

[itex]\begin{array}{ccc}<br /> a_{11}\: a_{12}\: a_{13} & & a_{21}\: a_{22}\: a_{23}\\<br /> \\<br /> 500 & & 300\\<br /> 050 & & 030\\<br /> 005 & & 003\\<br /> 410 & & 210\\<br /> 401 & & 201\\<br /> 140 & & 120\\<br /> 104 & & 102\\<br /> 041 & & 021\\<br /> 014 & & 012\\<br /> 320 & & 111\\<br /> 302\\<br /> 230\\<br /> 203\\<br /> 032\\<br /> 023\\<br /> 311\\<br /> 131\\<br /> 113\\<br /> 221\\<br /> 212\\<br /> 122<br /> \end{array}[/itex]

which has, as expected [itex]\binom{3+5-1}{5}\binom{3+3-1}{3}=21\cdot10[/itex] solutions.
Now let's consider also the vertical equations. For fixed [itex]\left(s_{1},s_{2},s_{3}\right)[/itex] some of those 210 solutions will be our final solutions. First of all in order to have solutions it must be [itex]s_{1}+s_{2}+s_{3}=5+3=8[/itex], so only the[itex]s[/itex] which satisfy this condition will contribute. For example, for [itex]\left(s_{1},s_{2},s_{3}\right)=\left(7,1,0\right)[/itex] from the previous table we select only

[itex]\begin{array}{ccc}<br /> a_{11}\: a_{12}\: a_{13} & & a_{21}\: a_{22}\: a_{23}\\<br /> \\<br /> 500 & & 210\\<br /> 410 & & 300<br /> \end{array}[/itex]

and thus we have two solutions. The number of solutions for all other possible [itex]\left(s_{1},s_{2},s_{3}\right)[/itex] are

[itex]\begin{array}{ccc}<br /> s_{1}\: s_{2}\: s_{3}\\<br /> 800 & \rightarrow & 1\\<br /> 710 & \rightarrow & 2\\<br /> 620 & \rightarrow & 3\\<br /> 611 & \rightarrow & 4\\<br /> 530 & \rightarrow & 4\\<br /> 521 & \rightarrow & 6\\<br /> 431 & \rightarrow & 7\\<br /> 422 & \rightarrow & 8\\<br /> 322 & \rightarrow & 9<br /> \end{array}[/itex]

(permutations of the [itex]s[/itex] give the same number of solutions).
How would one generalize all this? If [itex]\gamma[/itex] is the number of solutions of the general system, this identity is valid:

[itex]\sum_{s_{1}+..s_{n}=k_{1}+..k_{m}}\gamma\left(s_{1},..s_{n};k_{1},..k_{m}\right)=\prod_{i=1}^{m} \binom{n+k_{i}-1}{k_{i}}[/itex]

The point is to find that [itex]\gamma[/itex]...
 

Similar threads

Replies
4
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 14 ·
Replies
14
Views
2K