COCO: a platform for comparing continuous optimizers in a black-box setting

Hansen, Nikolaus; Auger, Anne; Ros, Raymond; Mersmann, Olaf; Tušar, Tea; Brockhoff, Dimo

doi:10.1080/10556788.2020.1808977

Cited by 275 publications

(184 citation statements)

References 52 publications

Supporting

Mentioning

157

Contrasting

Unclassified

Order By: Relevance

“…The difficulty of comparing alternative algorithm implementations and of assessing incremental changes to a given algorithm is now being increasingly recognized (e.g. Hansen et al, 2016;Sörensen, 2013;Weyland, 2012Weyland, , 2015. This has led some algorithm developers to formalize "standard" algorithm implementations that can be used to establish the expected baseline reference behaviour and performance of a given algorithm (Bratton and Kennedy, 2007;Swan et al, 2015).…”

Section: How Do We Implement Eas?mentioning

confidence: 99%

Introductory overview: Optimization using evolutionary algorithms and other metaheuristics

Maier

Razavi

Kapelan

et al. 2019

Environmental Modelling & Software

203

View full text Add to dashboard Cite

Environmental models are used extensively to evaluate the effectiveness of a range of design, planning, operational, management and policy options. However, the number of options that can be evaluated manually is generally limited, making it difficult to identify the most suitable options to consider in decision-making processes. By linking environmental models with evolutionary and other metaheuristic optimization algorithms, the decision options that make best use of scarce resources, achieve the best environmental outcomes for a given budget or provide the best tradeoffs between competing objectives can be identified. This Introductory Overview presents reasons for embedding formal optimization approaches in environmental decision-making processes, details how environmental problems are formulated as optimization problems and outlines how single-and multi-objective optimization approaches find good solutions to environmental problems. Practical guidance and potential challenges are also provided.

show abstract

Section: How Do We Implement Eas?mentioning

confidence: 99%

Introductory overview: Optimization using evolutionary algorithms and other metaheuristics

Maier

Razavi

Kapelan

et al. 2019

Environmental Modelling & Software

203

View full text Add to dashboard Cite

show abstract

“…function evaluations provided by computationally intensive simulations). [320] First benchmark, all functions shifted CEC'2011 [321] Real-world problems, small dimensions CEC'2013 [322] Rotated and shifted functions CEC'2014 [323] More multimodal functions CEC'2014 Expensive [324] Reduced number of evaluations CEC'2015 [325] Allowed specific parameter values for functions CEC'2017 [326] Composed test problems by extracting features dimension-wise for several problems BBOB [327] Functions with increasing dimensionality (from small dimensions)…”

Section: Benchmarks and Comparison Methodologiesmentioning

confidence: 99%

Bio-inspired computation: Where we stand and what's next

Ser

Osaba

Molina

et al. 2019

Swarm and Evolutionary Computation

481

106

View full text Add to dashboard Cite

show abstract

“…To check the utility of other, more efficient machine learning methods, we decided to include also linear regression, nearest neighbors and regression trees. For each of the 36 surrogate-relevator combinations, we tuned the five parameters of the meta model by using COCO, the platform for comparing numerical optimization methods in a black-box setting [8]. The parameters were tuned by using grid search with values as shown in Table 1.…”

Section: A Meta-model Tuning and Selectionmentioning

confidence: 99%

“…We select each algorithm in the pair among six alternative algorithms for learning predictive models, previously used in the literature on surrogate-based optimization-linear regression, decision trees, nearest neighbors, support vector machines, Gaussian processes and random forests-leading to 36 meta-model instances. In the first series of experiments, performed on synthetic benchmarks [8], we tune the parameters of each meta-model instance. In turn, we select the most successful instances that significantly outperform the others.…”

Section: Introductionmentioning

confidence: 99%

Meta-Model Framework for Surrogate-Based Parameter Estimation in Dynamical Systems

Lukšič¹,

Tanevski²,

Džeroski³

et al. 2019

IEEE Access

View full text Add to dashboard Cite

The central task in modeling complex dynamical systems is parameter estimation. This task is an optimization task that involves numerous evaluations of a computationally expensive objective function. Surrogate-based optimization introduces a computationally efficient predictive model that approximates the value of the objective function. The standard approach involves learning a surrogate from training examples that correspond to past evaluations of the objective function. Current surrogate-based optimization methods use static, predefined substitution strategies to decide when to use the surrogate and when the true objective. We introduce a meta-model framework where the substitution strategy is dynamically adapted to the solution space of the given optimization problem. The meta model encapsulates the objective function, the surrogate model and the model of the substitution strategy, as well as components for learning them. The framework can be seamlessly coupled with an arbitrary optimization algorithm without any modification: It replaces the objective function and autonomously decides how to evaluate a given candidate solution. We test the utility of the framework on three tasks of estimating parameters of real-world models of dynamical systems. The results show that the meta model significantly improves the efficiency of optimization, reducing the total number of evaluations of the objective function up to an average of 77%. INDEX TERMS Differential equations, meta models, numerical optimization, parameter estimation, surrogate models.

show abstract

COCO: a platform for comparing continuous optimizers in a black-box setting

Cited by 275 publications

References 52 publications

Introductory overview: Optimization using evolutionary algorithms and other metaheuristics

Introductory overview: Optimization using evolutionary algorithms and other metaheuristics

Bio-inspired computation: Where we stand and what's next

Meta-Model Framework for Surrogate-Based Parameter Estimation in Dynamical Systems

Contact Info

Product

Resources

About