Expected Improvement Matrix-Based Infill Criteria for Expensive Multiobjective Optimization

Zhan, Dawei; Cheng, Yuansheng; Li, Jun

doi:10.1109/tevc.2017.2697503

Cited by 153 publications

(41 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One such acquisition function is the expected improvement matrix (EIM) [28]. EIM is a multi-objective adaptation of EI, with the expected improvement for each candidate and each objective calculated in a matrix.…”

Section: Acquisition Functionmentioning

confidence: 99%

“…EIM has been previously implemented in the form of the EIM-EGO algorithm. The implementation was tested on a series of continuous test problems, displaying competitive results to the state-of-the-art multi-objective algorithms, whilst providing efficient scaling when increasing the number of optimised objectives [28].…”

Section: Acquisition Functionmentioning

confidence: 99%

“…The algorithm looks to extend Bayesian multi-objective methodologies to the mixed variable domain, which to the best of the authors' knowledge has limited prior work [27]. Utilising the recently proposed Expected Improvement Matrix [28] and an adapted distance metric, the algorithm provides an efficient approach to optimising expensive to evaluate mixed variable multiobjective optimisation problems without the need for reparameterization of the discrete variables.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MVMOO: Mixed variable multi-objective optimisation

2021

View full text Add to dashboard Cite

In many real-world problems there is often the requirement to optimise multiple conflicting objectives in an efficient manner. In such problems there can be the requirement to optimise a mixture of continuous and discrete variables. Herein, we propose a new multi-objective algorithm capable of optimising both continuous and discrete bounded variables in an efficient manner. The algorithm utilises Gaussian processes as surrogates in combination with a novel distance metric based upon Gower similarity. The MVMOO algorithm was compared to an existing mixed variable implementation of NSGA-II and random sampling for three test problems. MVMOO shows competitive performance on all proposed problems with efficient data acquisition and approximation of the Pareto fronts for the selected test problems.

show abstract

Section: Acquisition Functionmentioning

confidence: 99%

Section: Acquisition Functionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

MVMOO: Mixed variable multi-objective optimisation

2021

View full text Add to dashboard Cite

show abstract

“…where f * is the true value of the objective or fitness value of the solution, f is the approximated value, and ξ is the error function which reflects the degree of "uncertainty" of the approximation of the model [56].…”

Section: Single-objective Saeasmentioning

confidence: 99%

Model-based evolutionary algorithms: a short survey

Cheng

Jin

et al. 2018

Complex Intell. Syst.

View full text Add to dashboard Cite

The evolutionary algorithms (EAs) are a family of nature-inspired algorithms widely used for solving complex optimization problems. Since the operators (e.g. crossover, mutation, selection) in most traditional EAs are developed on the basis of fixed heuristic rules or strategies, they are unable to learn the structures or properties of the problems to be optimized. To equip the EAs with learning abilities, recently, various model-based evolutionary algorithms (MBEAs) have been proposed. This survey briefly reviews some representative MBEAs by considering three different motivations of using models. First, the most commonly seen motivation of using models is to estimate the distribution of the candidate solutions. Second, in evolutionary multi-objective optimization, one motivation of using models is to build the inverse models from the objective space to the decision space. Third, when solving computationally expensive problems, models can be used as surrogates of the fitness functions. Based on the review, some further discussions are also given.

show abstract

“…Despite those mentioned difficulties in real-world applications, many benchmark test suites, which try to mimic the properties of real-world problems, have been used to examine the performance of data-driven EMO algorithms. For instance, the KNO and OKA problems was used in [11]; the Zitzler-Deb-Thiele test suite (ZDT) [12] was used in [13][14][15][16]; the Deb-Thiele-Laumanns-Zitzler test suite (DTLZ) [17] was used in [18,19]; and the MF test suite was used in [20]. It is highlighted that these benchmark test suites promote the development of data-driven evolutionary multiobjective optimization, but the abilities of these data-driven EMO algorithms in solving real-world expensive MOPs are not validated.…”

Section: Introductionmentioning

confidence: 99%

A repository of real-world datasets for data-driven evolutionary multiobjective optimization

Tian

Wang

et al. 2019

Complex Intell. Syst.

View full text Add to dashboard Cite

Many real-world optimization applications have more than one objective, which are modeled as multiobjective optimization problems. Generally, those complex objective functions are approximated by expensive simulations rather than cheap analytic functions, which have been formulated as data-driven multiobjective optimization problems. The high computational costs of those problems pose great challenges to existing evolutionary multiobjective optimization algorithms. Unfortunately, there have not been any benchmark problems reflecting those challenges yet. Therefore, we carefully select seven benchmark multiobjective optimization problems from real-world applications, aiming to promote the research on data-driven evolutionary multiobjective optimization by suggesting a set of benchmark problems extracted from various real-world optimization applications.

show abstract

Expected Improvement Matrix-Based Infill Criteria for Expensive Multiobjective Optimization

Cited by 153 publications

References 41 publications

MVMOO: Mixed variable multi-objective optimisation

MVMOO: Mixed variable multi-objective optimisation

Model-based evolutionary algorithms: a short survey

A repository of real-world datasets for data-driven evolutionary multiobjective optimization

Contact Info

Product

Resources

About