Abstract — Tuning the parameters of an evolutionary algorithm (EA) to a given problem at hand is essential for good algorithm performance. Optimizing parameter values is, however, a non-trivial problem, beyond the limits of human problem solving.In this light it is odd that no parameter tuning algorithms are used widely in evolutionary computing. This paper is meant to be stepping stone towards a better practice by discussing the most important issues related to tuning EA parameters, describing a number of existing tuning methods, and presenting a modest experimental comparison among them. The paper is concluded by suggestions for future research – hopefully inspiring fellow researchers for further work. Index Terms — evolutionary algorithms, parameter tuning I. BACKGROUND AND OBJECTIVES Evolutionary Algorithms (EA) form a rich class of stochastic

Summary. This chapter presents a novel framework for tuning the parameters of Evolutionary Algorithms. A hybrid technique combining Meta-EAs and statistical Racing approaches is developed, which is not only capable of effectively exploring the search space of numerical parameters but also suitable for tuning symbolic parameters where it is generally difficult to define any sensible distance metric. 1

The research literature on metaheuristic and evolutionary computation has proposed a large number of algorithms for the solution of challenging real-world optimization problems. It is often not possible to study theoretically the performance of these algorithms unless significant assumptions are made on either the algorithm itself or the problems to which it is applied, or both. As a consequence, metaheuristics are typically evaluated empirically using a set of test problems. Unfortunately, relatively little attention has been given to the development of methodologies and tools for the large-scale empirical evaluation and/or comparison of metaheuristics. In this paper, we propose a landscape (test-problem) generator that can be used to generate optimization problem instances for continuous, bound-constrained optimization problems. The landscape generator is parameterized by a small number of parameters, and the values of these parameters have a direct and intuitive interpretation in terms of the geometric features of the landscapes that they produce. An experimental space is defined over algorithms and problems, via a tuple of parameters for any specified algorithm and problem class (here determined by the landscape generator). An experiment is then clearly specified as a point in this space, in a way that is analogous to other areas of experimental algorithmics, and more generally in experimental design. Experimental results are presented, demonstrating the use of the landscape generator. In particular, we analyze some simple, continuous estimation of distribution algorithms, and gain new insights into the behavior of these algorithms using the landscape generator.

Abstract-A pervasive problem in the field of optimization algorithms is the lack of meaningful and consistent algorithm benchmarking methodology. This includes but is not limited to issues of the selection of problem instances, the selection of algorithm specifications, the algorithm configuration parameters, and interpretation of results. The intention of this paper is to summarize the literature related to benchmarking optimization algorithms, with a focus on benchmarking in the face of the “no free lunch ” theorem, and useful statistical tools for interpreting results. This context for this review is biologically inspired optimization algorithms applied to continuous function optimization, although the principles extend beyond these themes.

Solving multi-objective optimization problems is a challenging task that demands efficient software tools and systematic analytical approaches. In this paper two evolutionary multi-objective optimization algorithms – namely the evolution strategy (ES) and the NSGA II – are applied to two complex real-world problems. The parameter settings of the evolutionary algorithms have been chosen and optimized according to statistical design plans. A new ranking method for measuring the quality of pareto-fronts is introduced. The layout of mold temperature control systems and the scheduling of elevators show typical complexity aspects that are necessary to illustrate a systematic approach of solving real-world multi-objective optimization problems. Keywords: Multi-objective evolutionary algorithms, Statistical design of experiments, Pareto-ranking method, Mold temperature control, Elevator supervisory group control. 1

Abstract- Choosing the best parameter setting is a wellknown important and challenging task in Evolutionary Algorithms (EAs). As one of the earliest parameter tuning techniques, the Meta-EA approach regards each parameter as a variable and the performance of algorithm as the fitness value and conducts searching on this landscape using various genetic operators. However, there are some inherent issues in this method. For example, some algorithm parameters are generally not searchable because it is difficult to define any sensible distance metric on them. In this paper, a novel approach is proposed by combining the Meta-EA approach with a method called Racing, which is based on the statistical analysis of algorithm performance with different parameter settings. A series of experiments are conducted to show the reliability and efficiency of this hybrid approach in tuning Genetic Algorithms (GAs) on two benchmark problems. 1

The success of evolutionary algorithms (EAs) depends crucially on finding suitable parameter settings. Doing this by hand is a very time consuming job without the guarantee to finally find satisfactory parameters. Of course, there exist various kinds of parameter control techniques, but not for parameter tuning. The Design of Experiment (DoE) paradigm offers a way of retrieving optimal parameter settings. It is still a tedious task, but it is known to be a robust and well tested suite, which can be beneficial for giving reason to parameter choices besides human experience. In this paper we analyse evolution strategies (ES) and particle swarm optimisation (PSO) with and without optimal parameters gathered with DoE. Reasonable improvements have been observed for the two ES variants.

Abstract. In practical applications, one can take advantage of metaheuristics in different ways: To simplify, we can say that metaheuristics can be either used out-of-the-box or a custom version can be developed. The former way requires a rather low effort, and in general allows to obtain fairly good results. The latter implies a larger investment in the design, implementation, and fine-tuning, and can often produce state-ofthe-art results. Unfortunately, most of the research works proposing an empirical analysis of metaheuristics do not even try to quantify the development effort devoted to the algorithms under consideration. In other words, they do not make clear whether they considered out-of-the-box or custom implementations of the metaheuristics under analysis. The lack of this information seriously undermines the generality and utility of these works. The aim of the paper is to stress that results obtained with out-ofthe-box implementations cannot be always generalized to custom ones, and vice versa. As a case study, we focus on the vehicle routing problem with stochastic demand and on five among the most successful metaheuristics—namely, tabu search, simulated annealing, genetic algorithm, iterated local search, and ant colony optimization. We show that the relative performance of these algorithms strongly varies whether one considers out-of-the-box implementations or custom ones, in which the parameters are accurately fine-tuned. 1

Abstract not found

Abstract- Sequential parameter optimization is a heuristic that combines classical and modern statistical techniques to improve the performance of search algorithms. To demonstrate its flexibility, three scenarios are discussed: (1) no experience how to choose the parameter setting of an algorithm is available, (2) a comparison with other algorithms is needed, and (3) an optimization algorithm has to be applied effectively and efficiently to a complex real-world optimization problem. Although sequential parameter optimization relies on enhanced statistical techniques such as design and analysis of computer experiments, it can be performed algorithmically and requires basically the specification of the relevant algorithm’s parameters. 1