Influence of ensemble surrogate models and sampling strategy on the solution quality of algorithms for computationally expensive black-box global optimization problems

Müller, Juliane; Shoemaker, Christine A.

doi:10.1007/s10898-014-0184-0

Cited by 144 publications

(95 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conejo et al (2013) develop a method which uses local quadratic models of the objective function for problems with a known convex feasible region. General constrained CDFO problems have been studied by Caballero and Grossmann (2008) and Sankaran et al (2010) using local kriging models for the objective and constraints; Powell (1994Powell ( , 2013a developed the COBYLA algorithm which uses local linear approximations for the unknown objective and the unknown constraints which is an approach revisited recently by March and Willcox (2012); Conn and Le Digabel (2013) employ quadratic models for the objective and constraints; Müller et al (2013) use radial-basis function models; and finally Müller and Shoemaker (2014) use an ensemble of various surrogate-models to best fit the objective and constraints of the problem. A recent extension of the BOBYQA algorithm for mixed variable programming has been published which uses quadratic approximations and local integer search, with guaranteed identification of locally optimal points (Newby and Ali, 2014).…”

Section: Model-based Methodsmentioning

confidence: 99%

Global optimization advances in Mixed-Integer Nonlinear Programming, MINLP, and Constrained Derivative-Free Optimization, CDFO

Boukouvala

Misener

Floudas

2016

European Journal of Operational Research

193

View full text Add to dashboard Cite

This manuscript reviews recent advances in deterministic global optimization for Mixed-Integer Nonlinear Programming (MINLP), as well as Constrained DerivativeFree Optimization (CDFO). This work provides a comprehensive and detailed literature review in terms of significant theoretical contributions, algorithmic developments, software implementations and applications for both MINLP and CDFO. Both research areas have experienced rapid growth, with a common aim to solve a wide range of real-world problems. We show their individual prerequisites, formulations and applicability, but also point out possible points of interaction in problems which contain hybrid characteristics. Finally, an inclusive and complete test suite is provided for both MINLP and CDFO algorithms, which is useful for future benchmarking.

show abstract

Section: Model-based Methodsmentioning

confidence: 99%

Global optimization advances in Mixed-Integer Nonlinear Programming, MINLP, and Constrained Derivative-Free Optimization, CDFO

Boukouvala

Misener

Floudas

2016

European Journal of Operational Research

193

View full text Add to dashboard Cite

show abstract

“…Müller and Piché (2011) introduced one such combination based on Dempster-Shafer theory, and Müller and Shoemaker (2014) parallelized it by adding random sampling similar to that of Regis and Shoemaker (2007b). RosalesPerez et al (2013) have experimented with an ensemble of SVM surrogates each with a different hyperparameters, and found good performance for the NSGA-II evolutionary multiobjective algorithm.…”

Section: Multiple Surrogatesmentioning

confidence: 99%

Parallel surrogate-assisted global optimization with expensive functions – a survey

Haftka

Villanueva

Chaudhuri

2016

Struct Multidisc Optim

204

View full text Add to dashboard Cite

Surrogate assisted global optimization is gaining popularity. Similarly, modern advances in computing power increasingly rely on parallelization rather than faster processors. This paper examines some of the methods used to take advantage of parallelization in surrogate based global optimization. A key issue focused on in this review is how different algorithms balance exploration and exploitation. Most of the papers surveyed are adaptive samplers that employ Gaussian Process or Kriging surrogates. These allow sophisticated approaches for balancing exploration and exploitation and even allow to develop algorithms with calculable rate of convergence as function of the number of parallel processors. In addition to optimization based on adaptive sampling, surrogate assisted parallel evolutionary algorithms are also surveyed. Beyond a review of the present state of the art, the paper also argues that methods that provide easy parallelization, like multiple parallel runs, or methods that rely on population of designs for diversity deserve more attention.

show abstract

“…Regis andShoemaker (2007, 2013) also use RBF models and new function evaluation points are selected by a stochastic method. Müller and Piché (2011) developed a framework for automatically computing ensembles of various surrogate model types and Müller and Shoemaker (2014) extended the study to investigate the influence of different sampling strategies on the solution quality. Here for the first time we apply a state-of-the-art RBF surrogate optimization algorithm to the problem of land surface emissions of methane and describe the results.…”

Section: Introductionmentioning

confidence: 99%

CH<sub>4</sub> parameter estimation in CLM4.5bgc using surrogate global optimization

et al. 2015

View full text Add to dashboard Cite

Abstract. Over the anthropocene methane has increased dramatically. Wetlands are one of the major sources of methane to the atmosphere, but the role of changes in wetland emissions is not well understood. The Community Land Model (CLM) of the Community Earth System Models contains a module to estimate methane emissions from natural wetlands and rice paddies. Our comparison of CH 4 emission observations at 16 sites around the planet reveals, however, that there are large discrepancies between the CLM predictions and the observations. The goal of our study is to adjust the model parameters in order to minimize the root mean squared error (RMSE) between model predictions and observations. These parameters have been selected based on a sensitivity analysis. Because of the cost associated with running the CLM simulation (15 to 30 min on the Yellowstone Supercomputing Facility), only relatively few simulations can be allowed in order to find a near-optimal solution within an acceptable time. Our results indicate that the parameter estimation problem has multiple local minima. Hence, we use a computationally efficient global optimization algorithm that uses a radial basis function (RBF) surrogate model to approximate the objective function. We use the information from the RBF to select parameter values that are most promising with respect to improving the objective function value. We show with pseudo data that our optimization algorithm is able to make excellent progress with respect to decreasing the RMSE. Using the true CH 4 emission observations for optimizing the parameters, we are able to significantly reduce the overall RMSE between observations and model predictions by about 50 %. The methane emission predictions of the CLM using the optimized parameters agree better with the observed methane emission data in northern and tropical latitudes. With the optimized parameters, the methane emission predictions are higher in northern latitudes than when the default parameters are used. For the tropics, the optimized parameters lead to lower emission predictions than the default parameters.

show abstract

Influence of ensemble surrogate models and sampling strategy on the solution quality of algorithms for computationally expensive black-box global optimization problems

Cited by 144 publications

References 43 publications

Global optimization advances in Mixed-Integer Nonlinear Programming, MINLP, and Constrained Derivative-Free Optimization, CDFO

Global optimization advances in Mixed-Integer Nonlinear Programming, MINLP, and Constrained Derivative-Free Optimization, CDFO

Parallel surrogate-assisted global optimization with expensive functions – a survey

CH<sub>4</sub> parameter estimation in CLM4.5bgc using surrogate global optimization

Contact Info

Product

Resources

About

Influence of ensemble surrogate models and sampling strategy on the solution quality of algorithms for computationally expensive black-box global optimization problems

Cited by 144 publications

References 43 publications

Global optimization advances in Mixed-Integer Nonlinear Programming, MINLP, and Constrained Derivative-Free Optimization, CDFO

Global optimization advances in Mixed-Integer Nonlinear Programming, MINLP, and Constrained Derivative-Free Optimization, CDFO

Parallel surrogate-assisted global optimization with expensive functions – a survey

CH&lt;sub&gt;4&lt;/sub&gt; parameter estimation in CLM4.5bgc using surrogate global optimization

Contact Info

Product

Resources

About

CH<sub>4</sub> parameter estimation in CLM4.5bgc using surrogate global optimization