On Symmetry Groups of Some Quadratic Programming Problems

TL;DR

This paper investigates the symmetry groups of quadratic programming problems with quadratic forms and positive definite constraints, focusing on orthogonal transformations to identify problem symmetries and reduce computational complexity.

Contribution

It extends symmetry analysis in mathematical programming to include orthogonal transformations, providing a method to find these symmetries in quadratic problems.

Findings

Characterization of symmetry subgroups under orthogonal transformations

Method for identifying symmetries demonstrated in examples

Potential for simplifying quadratic programming problems

Abstract

Solution and analysis of mathematical programming problems may be simplified when these problems are symmetric under appropriate linear transformations. In particular, a knowledge of the symmetries may help reduce the problem dimension, cut the search space by linear cuts or obtain new local optima from the ones previously found. While the previous studies of symmetries in the mathematical programming usually dealt with permutations of coordinates of the solutions space, the present paper considers a larger group of invertible linear transformations. We study a special case of the quadratic programming problem, where the objective function and constraints are given by quadratic forms, and the sum of all matrices of quadratic forms, involved in the constraints, is a positive definite matrix. In this setting, it is sufficient to consider only orthogonal transformations of the solution space. In this group of orthogonal transformations, we describe the structure of the subgroup which gives the symmetries of the problem. Besides that, a method for finding such symmetries is outlined, and illustrated in two simple examples.