Litcius/Paper detail

McSparse: Exact Solutions of Sparse Maximum Cut and Sparse Unconstrained Binary Quadratic Optimization Problems

Jonas Charfreitag, Michael Jünger, Sven Mallach, Petra Mutzel

2022Society for Industrial and Applied Mathematics eBooks15 citationsDOI

Abstract

While the Maximum Cut Problem and Unconstrained Binary Quadratic Optimization are of high interest in the scientific community and gain increasing importance, state-of-the-art solvers for these problems are publicly available, yet not for sparse instances of larger scale. We present the novel solver McSparse to fill this gap. It is installed as an internet service similar to the well-known services Biq Mac and BiqCrunch. We explain details of the algorithmic innovations based on integer linear programming and polyhedral combinatorics leading to the branch-and-cut algorithm implemented in McSparse. Substantial improvements with respect to former such approaches are demonstrated and the sustained performance is compared to those of other state-of-the-art methods using a broad set of benchmark instances.

Topics & Concepts

SolverBenchmark (surveying)Quadratic equationBinary numberQuadratic programmingSet (abstract data type)Computer scienceInteger programmingLinear programmingAlgorithmCode (set theory)MathematicsMathematical optimizationProgramming languageArithmeticGeodesyGeometryGeographyScheduling and Optimization AlgorithmsVehicle Routing Optimization MethodsOptimization and Search Problems
McSparse: Exact Solutions of Sparse Maximum Cut and Sparse Unconstrained Binary Quadratic Optimization Problems | Litcius