hal-01376611
https://enpc.hal.science/hal-01376611
https://enpc.hal.science/hal-01376611/document
https://enpc.hal.science/hal-01376611/file/lsmear-GOW-16.pdf
[ENPC] Ecole des Ponts ParisTech
[CNRS] CNRS - Centre national de la recherche scientifique
[LIGM_A3SI] Algorithms, architectures, image analysis and computer graphics
[PARISTECH] ParisTech
[LIGM] Laboratoire d'informatique Gaspard-Monge
[CV_LIGM] CV des membres du LIGM
[IMAGINE] IMAGINE
[ENPC_ONLY] ENPC ONLY
[ESIEE-PARIS] ESIEE Paris
[UNIV-EIFFEL] Université Gustave Eiffel
[UPEM-UNIVEIFFEL] Université Paris-Est Marne-la-Vallée
[ESIEE-UNIVEIFFEL] ESIEE Paris
[TEST3-HALCNRS] TEST3-HALCNRS
lsmear: A Variable Selection Strategy for Interval Branch and Bound Solvers
Araya, Ignacio
Neveu, Bertrand
[INFO.INFO-RO] Computer Science [cs]/Operations Research [cs.RO]
COMM
global optimization
branch and bound
variable selection
Lagrangian multipliers
smear function
Smear-based variable selection strategies are well-known and commonly used by branch-and-prune interval-based solvers. They estimates the impact of the variables on each constraint of the system by using the partial derivatives and the sizes of the variable domains. Then they aggregate these values, in some way, to estimate the impact of each variable on the whole system. The variable with the greatest impact is then selected. A problem of these strategies is that they, generally, consider all constraints equally important. In this work, we propose a new variable selection strategy which first weights the constraints by using the optimal Lagrangian multipliers of a linearization of the original problem. Then, the impact of the variables is computed with a typical smear-based function but taking into account the weights of the constraints. The strategy is tested on classical benchmark instances outperforming significantly the classical ones.
2016-09-04
2016-10-05
en
GOW'16 XIII Global Optimization Workshop
Braga, Portugal