Welcome
TwitterFacebookGoogle

A CMA-ES-based 2-stage memetic framework for solving constrained optimization problems

Constraint optimization problems play a crucial role in many application domains, ranging from engineering design to finance and logistics. Specific techniques are therefore needed to handle complex fitness landscapes characterized by multiple constraints. In the last decades, a number of novel meta-heuristics have been applied to constraint optimization. Among these, the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) has been attracting lately the most attention of researchers. Recent variants of CMA-ES showed promising results on several benchmarks and practical problems. In this paper, we attempt to improve the performance of an adaptive penalty CMA-ES recently proposed in the literature. We build upon it a 2-stage memetic framework, coupling the CMA-ES scheme with a local optimizer, so that the best solution found by CMA-ES is used as starting point for the local search. We test, separately, the use of three classic local search algorithms (Simplex, BOBYQA, and L-BFGS-B), and we compare the baseline scheme (without local search) and its three memetic variants with some of the state-of-the-art methods for constrained optimization.

 

Highlights:

  • We introduce a modified Covariance Matrix Adaptation Evolution Strategy.
  • Our CMA-ES is specifically designed for solving constrained optimization problems.
  • The proposed method is a 2-stage memetic.
  • Several local-search methods were tested.
  • This novel CMA-ES presents competitive results on a set of benchmark functions.

Veloso de Melo, Vinicius; Iacca, Giovanni, “A CMA-ES-based 2-stage memetic framework for solving constrained optimization problems,” Foundations of Computational Intelligence (FOCI), 2014 IEEE Symposium on , vol., no., pp.143,150, 9-12 Dec. 2014
doi: 10.1109/FOCI.2014.7007819

 

@INPROCEEDINGS{7007819,
author={Veloso de Melo, Vinicius and Iacca, Giovanni},
booktitle={Foundations of Computational Intelligence (FOCI), 2014 IEEE Symposium on},
title={A CMA-ES-based 2-stage memetic framework for solving constrained optimization problems},
year={2014},
month={Dec},
pages={143-150},
keywords={Approximation methods;Covariance matrices;Memetics;Optimization;Search problems;Sociology},
doi={10.1109/FOCI.2014.7007819},}

 
Keywords:
Optimització constret. Matriu de covariància adaptació evolució estratègia. Meméticas. Cerca local.
约束的优化问题。协方差矩阵适应进化策略。模因。本地搜索。
約束的優化問題。共變數矩陣適應進化策略。模因。本地搜索。
Optimisation contrainte. Stratégie d’évolution Adaptation matrice de covariance. Memetic. Recherche locale.
Eingeschränkte Optimierung. Kovarianz Matrix Evolution Anpassungsstrategie. Memetic. Lokale Suche.
विवश अनुकूलन। सहप्रसरण मैट्रिक्स अनुकूलन विकास रणनीति। Memetic. स्थानीय खोज।
Dihalang pengoptimuman. Covariance matriks strategi evolusi adaptasi. Memetic. Carian tempatan.
Optimización con restricciones. Covarianza matriz estrategia de evolución de adaptación. Memética. Búsqueda local.