Parameter Sweeps for Exploring Parameter Spaces of Genetic and Evolutionary Algorithms

Samples, Michael E.; Byom, Matthew J.; Daida, Jason M.

doi:10.1007/978-3-540-69432-8_8

Cited by 9 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Antibiotic resistance evolves and disseminates in a complex parameter space. This space is determined by a limited set of axes determining all possible combinations of values for all different parameters ( 77 ). Different regions of the parameter space produce different types of local behaviour, expressed as families of probability distributions.…”

Section: A Space Of Composite Parameters Determining the Emergence Anmentioning

confidence: 99%

Emergence and spread of antibiotic resistance: setting a parameter space

Martínez

Baquero

2014

Upsala Journal of Medical Sciences

151

View full text Add to dashboard Cite

The emergence and spread of antibiotic resistance among human pathogens is a relevant problem for human health and one of the few evolution processes amenable to experimental studies. In the present review, we discuss some basic aspects of antibiotic resistance, including mechanisms of resistance, origin of resistance genes, and bottlenecks that modulate the acquisition and spread of antibiotic resistance among human pathogens. In addition, we analyse several parameters that modulate the evolution landscape of antibiotic resistance. Learning why some resistance mechanisms emerge but do not evolve after a first burst, whereas others can spread over the entire world very rapidly, mimicking a chain reaction, is important for predicting the evolution, and relevance for human health, of a given mechanism of resistance. Because of this, we propose that the emergence and spread of antibiotic resistance can only be understood in a multi-parameter space. Measuring the effect on antibiotic resistance of parameters such as contact rates, transfer rates, integration rates, replication rates, diversification rates, and selection rates, for different genes and organisms, growing under different conditions in distinct ecosystems, will allow for a better prediction of antibiotic resistance and possibilities of focused interventions.

show abstract

Section: A Space Of Composite Parameters Determining the Emergence Anmentioning

confidence: 99%

Emergence and spread of antibiotic resistance: setting a parameter space

Martínez

Baquero

2014

Upsala Journal of Medical Sciences

151

View full text Add to dashboard Cite

show abstract

“…Even if one performs a full parameter sweep over thousands of different parameters settings, the resulting entropy is still just an estimation. Although such a sweep can be distributed over multiple machines [30], it is still a very time consuming task. Especially because much time is spent on evaluating parameter settings that are not interesting, because their performance is far from optimal.…”

Section: Estimating Entropymentioning

confidence: 99%

Using Entropy for Parameter Analysis of Evolutionary Algorithms

Smit

Eiben

2010

Experimental Methods for the Analysis of Optimization Algorithms

View full text Add to dashboard Cite

Abstract-Evolutionary Algorithms (EA) form a rich class of stochastic search methods that share the basic principles of incrementally improving the quality of a set of candidate solutions by means of variation and selection [10], [8]. Such variation and selection operators often require parameters to be specified. Finding a good set of parameter values is a nontrivial problem in itself, furthermore some EA parameters are more relevant than others in the sense that choosing different values for them affects EA performance more than for the other parameters. In this chapter we explain the notion of entropy and discuss how entropy can disclose important information on EA parameters, in particular, about their relevance. We describe an algorithm that is able to estimate the entropy of EA parameters and we present a showcase, based on extensive experimentation, to demonstrate the usefulness of this approach and some interesting insights that are gained.

show abstract

“…Throughout the relevant literature we find that the cost of tuning parameters is largely ignored. Notable exceptions are the theoretical considerations of [17] and [9], as well as the systematic parameter sweeps of [11,21,20] and the statistical analysis of parameters by [6]. In the general field of experimental design, a paradigm shift that emphasizes a low cost of tuning over the performance of optimal parameter values was due to [22].…”

Section: Introductionmentioning

confidence: 99%

“…In our field, [7] proposes a meta-GA approach in which both EA components and EA parameters are tuned and shows the importance of the right choice for the GA operators. [20] shows how parameter sweeps can be used for robustness and correlation analysis. [18] embed sequential parameter optimization in a wider framework of experimental EA design.…”

Section: Introductionmentioning

confidence: 99%

Costs and Benefits of Tuning Parameters of Evolutionary Algorithms

Nannen

Smit

Eiben

2008

Parallel Problem Solving From Nature – PPSN X

View full text Add to dashboard Cite

Abstract. We present an empirical study on the impact of different design choices on the performance of an evolutionary algorithm (EA). Four EA components are considered-parent selection, survivor selection, recombination and mutation-and for each component we study the impact of choosing the right operator and of tuning its free parameter(s). We tune 120 different combinations of EA operators to 4 different classes of fitness landscapes and measure the cost of tuning. We find that components differ greatly in importance. Typically the choice of operator for parent selection has the greatest impact, and mutation needs the most tuning. Regarding individual EAs however, the impact of design choices for one component depends on the choices for other components, as well as on the available amount of resources for tuning.

show abstract

Parameter Sweeps for Exploring Parameter Spaces of Genetic and Evolutionary Algorithms

Cited by 9 publications

References 14 publications

Emergence and spread of antibiotic resistance: setting a parameter space

Emergence and spread of antibiotic resistance: setting a parameter space

Using Entropy for Parameter Analysis of Evolutionary Algorithms

Costs and Benefits of Tuning Parameters of Evolutionary Algorithms

Contact Info

Product

Resources

About