Multi‐objective ensemble generation

Gu, Shengfeng; Cheng, Ran; Jin, Yaochu

doi:10.1002/widm.1158

Cited by 32 publications

(18 citation statements)

References 91 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These techniques are categorized into explicit and implicit methods depending on whether optimization of a diversity metric is applied in ensemble construction [53]. Most of these techniques belong to the implicit type, such as training data manipulation including bagging [34], [54], feature selection and feature extraction [55], or use of different initial parameters, different training algorithms and different architectures of learners [56], while boosting [57] and negative correlation learning [58] belong to the explicit type. Heterogeneous ensembles have the natural advantages in structural diversity by using different architectures or different learning types of models [42], [55], [59].…”

Section: Heterogeneous Ensemble Generationmentioning

confidence: 99%

Heterogeneous Ensemble-Based Infill Criterion for Evolutionary Multiobjective Optimization of Expensive Problems

Guo

Jin

Ding

et al. 2019

IEEE Trans. Cybern.

Self Cite

177

View full text Add to dashboard Cite

Gaussian processes (GPs) are the most popular model used in surrogate-assisted evolutionary optimization of computationally expensive problems, mainly because GPs are able to measure the uncertainty of the estimated fitness values, based on which certain infill sampling criteria can be used to guide the search and update the surrogate model. However, the computation time for constructing GPs may become excessively long when the number of training samples increases, which makes it inappropriate to use them as surrogates in evolutionary optimization. To address this issue, this paper proposes to use ensembles as surrogates and infill criteria for model management in evolutionary optimization. A heterogeneous ensemble consisting of a least square support vector machine and two radial basis function networks is constructed to enhance the reliability of ensembles for uncertainty estimation. In addition to the original decision variables, a selected subset of the decision variables and a set of transformed variables are used as inputs of the heterogeneous ensemble to further promote the diversity of the ensemble. The proposed heterogeneous ensemble is compared with a GP and a homogeneous ensemble for infill sampling criteria in evolutionary multiobjective optimization. Experimental results demonstrate that the heterogeneous ensemble is competitive in performance compared with GPs and much more scalable in computational complexity to the increase in search dimension.

show abstract

Section: Heterogeneous Ensemble Generationmentioning

confidence: 99%

Heterogeneous Ensemble-Based Infill Criterion for Evolutionary Multiobjective Optimization of Expensive Problems

Guo

Jin

Ding

et al. 2019

IEEE Trans. Cybern.

Self Cite

177

View full text Add to dashboard Cite

show abstract

“…This can be done by the manipulation of: features; training instances; and/or model's hyperparameters. According to Gu et al (2015), the formulation of a MOP for finding the trade-off between prediction scores and complexity is presented as the most promising such task.…”

Section: Multi-objective Ensemble Generationmentioning

confidence: 99%

“…The application of MOO algorithms for the optimisation of ensemble learning models has been performed by many authors, and is known as multi-objective ensemble generation (MOEG) (Gu et al, 2015). Most of the works deal with ensemble member generation, where MOO is applied in order to generate a set of diverse predictive models (Tan et al, 2014;Smith and Jin, 2014;Rosales-Pérez et al, 2014;Gu and Jin, 2014;Rosales-Pérez et al, 2017;Peimankar et al, 2017;Ribeiro and Reynoso-Meza, 2018).…”

Section: Introductionmentioning

confidence: 99%

A Multi-objective Optimisation Design Approach for Ensemble Member Combination in Binary Classification Tasks

Ribeiro

Coelho

et al. 2019

Anais Do 14º Simpósio Brasileiro De Automação Inteligente

View full text Add to dashboard Cite

Machine learning algorithms are usually trained by means of optimising a single global score, and it is possible to have a poor performance when dealing with classification problems where a lower type I or type II error rate is more desirable. To deal with this situation, multi-objective optimisation techniques can be applied for finding the trade-off among the different objectives. However, such approach results in a set of non-dominated solutions, and it is important to select one preferable solution. Thus, this paper proposes the application of a holistic multi-objective optimisation design procedure for ensemble member combination. To do so, a multi-objective problem is formulated in order to maximise the class-specific recalls of a weighted majority voting ensemble. Such problem is optimised with a multi-objective evolutionary algorithm. Next, a ranking technique is used find a preferred solution based on the during the multi-criteria decision making step.

show abstract

“…There could be scope for research in determining time-related behavior in textile data. 72 Contributions are also expected on the topic of classifying textile data using ensemble learning 73 techniques in the coming years. In the textile industry, a single classifier has been used so far; however, combining multiple models can provide a more accurate prediction than depending on a single model.…”

Section: Future Directionsmentioning

confidence: 99%

Data mining and machine learning in textile industry

Yıldırım

Birant

Alpyıldız

2017

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Data mining has been proven useful for knowledge discovery in many areas, ranging from marketing to medical and from banking to education. This study focuses on data mining and machine learning in textile industry as applying them to textile data is considered an emerging interdisciplinary research field. Thus, data mining studies, including classification and clustering techniques and machine learning algorithms, implemented in textile industry were presented and explained in detail in this study to provide an overview of how clustering and classification techniques can be applied in the textile industry to deal with different problems where traditional methods are not useful. This article clearly shows that a classification technique has higher interest than a clustering technique in the textile industry. It also shows that the most commonly applied classification methods are artificial neural networks and support vector machines, and they generally provide high accuracy rates in the textile applications. For the clustering task of data mining, a K-means algorithm was generally implemented in textile studies among the others that were investigated in this article. We conclude with some remarks on the strength of the data mining techniques for textile industry, ways to overcome certain challenges, and offer some possible further research directions.

show abstract

Multi‐objective ensemble generation

Cited by 32 publications

References 91 publications

Heterogeneous Ensemble-Based Infill Criterion for Evolutionary Multiobjective Optimization of Expensive Problems

Heterogeneous Ensemble-Based Infill Criterion for Evolutionary Multiobjective Optimization of Expensive Problems

A Multi-objective Optimisation Design Approach for Ensemble Member Combination in Binary Classification Tasks

Data mining and machine learning in textile industry

Contact Info

Product

Resources

About