This is the tool for automatic dominance breaking for MiniZinc modelling language.
You can use MiniZinc to model constraint satisfaction and optimisation problems in a high-level, solver-independent way. The tool will analyze the model and generate dominannce breaking constraints if there are any.
To get a MiniZinc up and running follow these simple steps.
- CMake (>=3.4)
- A recent C++ compiler - Compilation is tested with recent versions of Clang, GCC, and Microsoft Visual C++.
Use the following command to clone the repo and the submodules
git clone --recurse-submodules https://github.com/AllenZzw/auto-dom.git
Then, enter the folder auto-dom and use the following command to compile the tool.
bash build.sh
The package has been tested on Linux and Mac environment.
Once the tool is built on your machine, you can find a binary named "auto-dom" in the "build" directory.
You can start using the tool to analyze your discrete optimisation models.
The following code segment shows a MiniZinc model for the well known Knapsack problem.
int: n = 10; % number of items
int: W = 40; % weight limit
array [1..n] of int: w = [5, 7, 8, 3, 12, 9, 5, 6, 7, 4]; % weight of each item
array [1..n] of int: v = [3, 11, 5, 9, 11, 8, 13, 11, 2, 15]; % value of each item
array [1..n] of var 0..1: x; % take item or not
constraint sum (i in 1..n) (w[i]*x[i]) <= W;
solve maximize sum (i in 1..n) (v[i]*x[i]);
Using the following command
build/minizinc knapsack.mzn
The output will be the dominance breaking constraints for the knapsack problem like
constraint x[1] != 1 \/ x[4] != 0;
constraint x[1] != 1 \/ x[7] != 0;
constraint x[1] != 0 \/ x[9] != 1;
constraint x[1] != 1 \/ x[10] != 0;
constraint x[2] != 0 \/ x[3] != 1;
constraint x[2] != 0 \/ x[6] != 1;
constraint x[2] != 1 \/ x[7] != 0;
constraint x[2] != 1 \/ x[8] != 0;
constraint x[2] != 0 \/ x[9] != 1;
constraint x[2] != 1 \/ x[10] != 0;
...
More details can be found in the follwing paper:
@inproceedings{jlee2022exploiting,
title={Exploiting Functional Constraints in Generating Dominance Breaking Nogoods for Constraint Optimization},
author={Jimmy H.M. Lee and Allen Z. Zhong},
booktitle={Proceedings of the 28th International Conference on Principles and Practice of Constraint Programming},
year={2022},
organization={Schloss Dagstuhl-Leibniz-Zentrum f{\"u}r Informatik}
}