Back to the program.
Memetic Algorithms for Continuous Optimisation
Daniel Molina
Univeristy of Cadiz, Dept. of Computer Science and Engineering
daniel.molina@uca.es

Abstract

Many real-world problems may be formulated as optimisation problems of parameters with variables in continuous domains (i.e., continuous optimisation problems). Over the past few years, an increasing interest has arisen in solving this kind of problem using different EA models, and different algorithms, like CMA-ES, with a high exploration capacity. In this field, memetic algorithms arise as very effective algorithms to obtain reliable and high accurate solutions for complex continuous optimization problems. Given the potential of these local optimisation methods, it is interesting to build prospective memetic algorithm models with them. However, they are can be very expensive, because of the way they exploit local information to guide the search process. In this talk I introduce the concept of local search chain as a springboard to design memetic algorithm approaches that can effectively use intense continuous local search methods as local search operators. Local search chain concerns the idea that, at one stage, the local search operator may continue the operation of a previous invocation, starting from the final configuration (initial solution, strategy parameter values, internal variables, etc.) reached by this one. The proposed memetic algorithm favours the formation of local search chains during the memetic algorithm run with the aim of concentrating local tuning in search regions showing promise. This model has been successfully used to create algorithms, MA-LSCh-CMA, with a good behaviour, in comparisons with a standard benchmarks (proposed in CEC'2005). Also, this model has been used with other local search methods to optimise high-dimensional numeric optimisation problems, obtaining the best place in a special session competition in CEC'2010. Also, I'll introduce the current state of continuous optimisations showing EAs that have obtained the best results in the last years, over a set of well know benchmarks and competitions.

Keywords

Memetic Algorithms, Continuous Optimisation, adaptive local search intensity

References

  1. Daniel Molina and Manuel Lozano and Carlos García-Martínez and Francisco Herrera. (2011) Memetic Algorithms for Continuous Optimisation Based on Local Search Chains, Evolutionary computation, 18(1):27-63. Other publications available on my website.
    See http://sci2s.ugr.es/publications/byAll.php?author=0016 & type=01 & name=International%20Journal%20Papers%20by%20D.%A0Molina