Model-based automatic neighborhood design by unsupervised learning |
| |
Affiliation: | 1. Commissariat à l’Energie Atomique, Direction de l’Energie Nucléaire, Laboratoire d’Etude des Matériaux Actifs, Bagnols sur Cèze BP 17171, 30207, France;2. Synchrotron SOLEIL, L’Orme des Merisiers, Saint Aubin, BP 48, F-91192 Gif-sur-Yvette Cedex, France;1. Innovative Technology research Center, AGC Inc., 1150 Hazawa-cho, Yokohama 221-8755, Japan;2. University College London, Gower Street, London W1E6BT, UK |
| |
Abstract: | The definition of a suitable neighborhood structure on the solution space is a key step when designing a heuristic for Mixed Integer Programming (MIP). In this paper, we move on from a MIP compact formulation and show how to take advantage of its features to automatically design efficient neighborhoods, without any human analysis. In particular, we use unsupervised learning to automatically identify “good” regions of the search space “around” a given feasible solution. Computational results on compact formulations of three well-known combinatorial optimization problems show that, on large instances, the neighborhoods constructed by our procedure outperform state-of-the-art domain-independent neighborhoods. |
| |
Keywords: | Mixed integer programming Neighborhood search Unsupervised learning |
本文献已被 ScienceDirect 等数据库收录! |
|