Salvador: Approximate Vertex Cover Solver

Current version: v0.1.3

This work builds upon The Salvador Algorithm.

The Minimum Vertex Cover Problem

The Minimum Vertex Cover (MVC) problem is a classic optimization problem in computer science and graph theory. It involves finding the smallest set of vertices in a graph that covers all edges, meaning at least one endpoint of every edge is included in the set.

Formal Definition

Given an undirected graph $G = (V, E)$, a vertex cover is a subset $V' \subseteq V$ such that for every edge $(u, v) \in E$, at least one of $u$ or $v$ belongs to $V'$. The MVC problem seeks the vertex cover with the smallest cardinality.

Importance and Applications

• Theoretical Significance: MVC is a well-known NP-hard problem, central to complexity theory.
• Practical Applications:
  • Network Security: Identifying critical nodes to disrupt connections.
  • Bioinformatics: Analyzing gene regulatory networks.
  • Wireless Sensor Networks: Optimizing sensor coverage.

Related Problems

• Maximum Independent Set: The complement of a vertex cover.
• Set Cover Problem: A generalization of MVC.

Problem Statement

Input: A Boolean Adjacency Matrix $M$.

Answer: Find a Minimum Vertex Cover.

Example Instance: 5 x 5 matrix

The input for undirected graph is typically provided in DIMACS format. In this way, the previous adjacency matrix is represented in a text file using the following string representation:

p edge 5 4
e 1 3
e 1 5
e 2 4
e 3 5

This represents a 5x5 matrix in DIMACS format such that each edge $(v,w)$ appears exactly once in the input file and is not repeated as $(w,v)$. In this format, every edge appears in the form of

e W V

where the fields W and V specify the endpoints of the edge while the lower-case character e signifies that this is an edge descriptor line.

Example Solution:

Vertex Cover Found 1, 3, 4: Nodes 1, 3, and 4 constitute an optimal solution.

Compile and Environment

Prerequisites

• Python ≥ 3.12

Installation

pip install salvador==0.1.3

Execution

1. Clone the repository:

   git clone https://github.com/frankvegadelgado/salvador.git
   cd salvador
   

2. Run the script:

   vega -i ./benchmarks/testMatrix1
   

   utilizing the vega command provided by Salvador to execute the Boolean adjacency matrix salvador\benchmarks\testMatrix1. The file testMatrix1 represents the example described herein. We also support .xz, .lzma, .bz2, and .bzip2 compressed text files.

   Example Output:

   testMatrix1: Vertex Cover Found 1, 3, 4
   

   This indicates nodes 1, 3, 4 form a vertex cover.

Vertex Cover Size

Use the -c flag to count the nodes in the vertex cover:

vega -i ./benchmarks/testMatrix2 -c

Output:

testMatrix2: Vertex Cover Size 5

Command Options

Display help and options:

vega -h

Output:

usage: vega [-h] -i INPUTFILE [-a] [-b] [-c] [-v] [-l] [--version]

Compute the Approximate Vertex Cover for undirected graph encoded in DIMACS format.

options:
  -h, --help            show this help message and exit
  -i INPUTFILE, --inputFile INPUTFILE
                        input file path
  -a, --approximation   enable comparison with a polynomial-time approximation approach within a factor of at most 2
  -b, --bruteForce      enable comparison with the exponential-time brute-force approach
  -c, --count           calculate the size of the vertex cover
  -v, --verbose         enable verbose output
  -l, --log             enable file logging
  --version             show program's version number and exit

Batch Execution

Batch execution allows you to solve multiple graphs within a directory consecutively.

To view available command-line options for the batch_vega command, use the following in your terminal or command prompt:

batch_vega -h

This will display the following help information:

usage: batch_vega [-h] -i INPUTDIRECTORY [-a] [-b] [-c] [-v] [-l] [--version]

Compute the Approximate Vertex Cover for all undirected graphs encoded in DIMACS format and stored in a directory.

options:
  -h, --help            show this help message and exit
  -i INPUTDIRECTORY, --inputDirectory INPUTDIRECTORY
                        Input directory path
  -a, --approximation   enable comparison with a polynomial-time approximation approach within a factor of at most 2
  -b, --bruteForce      enable comparison with the exponential-time brute-force approach
  -c, --count           calculate the size of the vertex cover
  -v, --verbose         enable verbose output
  -l, --log             enable file logging
  --version             show program's version number and exit

Testing Application

A command-line utility named test_vega is provided for evaluating the Algorithm using randomly generated, large sparse matrices. It supports the following options:

usage: test_vega [-h] -d DIMENSION [-n NUM_TESTS] [-s SPARSITY] [-a] [-b] [-c] [-w] [-v] [-l] [--version]

The Salvador Testing Application using randomly generated, large sparse matrices.

options:
  -h, --help            show this help message and exit
  -d DIMENSION, --dimension DIMENSION
                        an integer specifying the dimensions of the square matrices
  -n NUM_TESTS, --num_tests NUM_TESTS
                        an integer specifying the number of tests to run
  -s SPARSITY, --sparsity SPARSITY
                        sparsity of the matrices (0.0 for dense, close to 1.0 for very sparse)
  -a, --approximation   enable comparison with a polynomial-time approximation approach within a factor of at most 2
  -b, --bruteForce      enable comparison with the exponential-time brute-force approach
  -c, --count           calculate the size of the vertex cover
  -w, --write           write the generated random matrix to a file in the current directory
  -v, --verbose         enable verbose output
  -l, --log             enable file logging
  --version             show program's version number and exit

Reproducibility Experiments (car/)

The car/ folder contains the experiment scripts. Each script writes its results next to itself as .json and .csv files.

How to run

Recommended: run every experiment quickly, on small graphs, with one command from the car/ folder:

cd car
python run_all.py --quick

Install the package and its dependencies, then run a script from the repository root:

pip install salvador==0.1.3
python car/car_strategies.py

The scripts also run from a clone of the repository without installing the package. Every script accepts -h to list its options, --quick for a shorter run and --timeout to limit the time spent on each graph.

To run all the experiments one after another, use car/run_all.py:

python car/run_all.py --quick                          # every experiment on small graphs (much shorter)
python car/run_all.py --quick --only strategies scaling # only some experiments
python car/run_all.py --list                           # list the experiments and their quick options
python car/run_all.py                                  # every experiment in full (hours)
python car/run_all.py --skip triangle_free mechanism   # full runs except some

Each experiment writes its own result files, and car/run_all_log.txt records the time and status of each one.

What each script does

Examples:

python car/car_ratio.py
python car/car_bipartite.py
python car/car_scaling.py --quick
python car/car_triangle_free.py --quick
python car/car_c0_fix.py                                # full c0 study (about 35 minutes)

Code

• Python implementation by Frank Vega.

License

• MIT License.