|
 
|
[EL17a] Quadratic convex reformulation for partitioning problems
Conférence Internationale avec comité de lecture : EUROPT 17, July 2017, pp.1--1, Montreal, Canada,Mots clés: QAP, QCR, experiments
Résumé:
Quadratic convex reformulation is a general method to solve quadratic programs. We consider its specialization to partitioning-like problems, namely, quadratic assignment and k-way graph partitioning. We compare two reformulation families. In the first one, we rely on the space of the initial variables only and obtain compact reformulations. In the second family, we aim at getting tighter relaxation bounds by increasing the size of the reformulated problem. We illustrate through experiments.
BibTeX
| @inproceedings { | |||
| EL17a, | |||
| title | = | "{Quadratic convex reformulation for partitioning problems}", | |
| author | = | " S. Elloumi and A. Lambert ", | |
| booktitle | = | "{EUROPT 17}", | |
| year | = | 2017, | |
| month | = | "July", | |
| pages | = | "1--1", | |
| address | = | "Montreal, Canada", | |
| } | |||
