Solving combinatorial bi-level optimization problems using multiple populations and migration schemes

Said, Rihab; Elarbi, Maha; Bechikh, Slim; Saïd, Lamjed Ben

doi:10.1007/s12351-020-00616-z

Cited by 9 publications

(6 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The scope of bi-level multiobjective problems is vast and has been constantly growing during the past years. This is pointed out in the extensive reviews of Lachhwani and Dwivedi (2018), Sinha et al (2018) and Said et al (2021). Bilevel optimization problems arise in many practical settings.…”

Section: Bi-level Optimization and Price Coordinationmentioning

confidence: 98%

Integration of production scheduling and energy-cost optimization using Mean Value Cross Decomposition

Hadera

Ekström

Sand

et al. 2019

Computers & Chemical Engineering

View full text Add to dashboard Cite

Section: Bi-level Optimization and Price Coordinationmentioning

confidence: 98%

Integration of production scheduling and energy-cost optimization using Mean Value Cross Decomposition

Hadera

Ekström

Sand

et al. 2019

Computers & Chemical Engineering

View full text Add to dashboard Cite

“…In other words, the approximation of optimal lower level solutions requires a high number of evaluations. To deal with this high computational cost and to solve the resulting bilevel discretization-based feature selection problem, we have designed an improved version of the CEMBA [23], called I-CEMBA, that ensures the variation of the number of features during the migration step. The main goal behind proposing I-CEMBA is to tackle the multimodality aspect caused by having several feature subsets with the same number of features.…”

Section: Proposed Approach a Main Idea And Motivationsmentioning

confidence: 99%

“…2). 2) Solving the proposed bi-level model using an improved version of the bi-level algorithm CEMBA (Co-Evolutionary Migration-Based Algorithm) [23] which we name I-CEMBA, through the design of a new migration strategy. 3) Showing the ability of Bi-DFS in obtaining better results in terms of classification accuracy, generalization ability, and feature selection bias compared to classical and recent evolutionary approaches.…”

Section: Introductionmentioning

confidence: 99%

Discretization-Based Feature Selection as a Bilevel Optimization Problem

Said

Elarbi

Bechikh

et al. 2023

IEEE Trans. Evol. Computat.

Self Cite

View full text Add to dashboard Cite

Discretization-based feature selection approaches have shown interesting results when using several metaheuristic algorithms such as Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Ant Colony Optimization (ACO), etc. However, these methods share the same shortcoming which consists in encoding the problem solution as a sequence of cut-points. From this cut-points vector, the decision of deleting or selecting any feature is induced. Indeed, the number of generated cutpoints varies from one feature to another. Thus, the higher the number of cut-points, the higher the probability of selecting the considered feature and vice versa. This fact leads to the deletion of possibly important features having a single or a low number of cut-points, such as the infection rate, the glycemia level, and the blood pressure. In order to solve the issue of the dependency relation between the feature selection (or removal) event and the number of its generated potential cut-points, we propose to model the discretization-based feature selection task as a bi-level optimization problem and then solve it using an improved version of an existing co-evolutionary algorithm, named I-CEMBA. The latter ensures the variation of the number of features during the migration process in order to deal with the multimodality aspect. The resulting algorithm, termed Bi-DFS (Bi-level Discretization-based Feature Selection), performs selection at the upper level while discretization is done at the lower level. The experimental results on several high-dimensional datasets show that Bi-DFS outperforms relevant state-of-the-art methods in terms of classification accuracy, generalization ability, and feature selection bias.

show abstract

“…It is evaluated to be good if the determination of x 2 results in generating a good solution. For generating the good solution, the value of x 2 is determined by applying metaheuristics or greedy algorithms [2][3][4][5][6][7]. However, the metaheuristics are timeconsuming, and the greedy algorithms are not general-purpose.…”

Section: Basic Conceptmentioning

confidence: 99%

“…When a GA is applied to an optimization problem, especially to one with two decision variable vectors, the following strategy is often adopted [2][3][4][5][6][7]. An individual of GA is expressed by the values of one of the decision variable vectors.…”

mentioning

confidence: 99%

Genetic Algorithm with Machine Learning to Estimate the Optimal Objective Function Values of Subproblems

Iima

Hazama

2022

Proceedings of the 2022 6th International Conference on Intelligent Systems, Metaheuristics &Amp; Swarm Intelligence

View full text Add to dashboard Cite

This paper addresses an optimization problem with two decision variable vectors. This problem can be divided into multiple subproblems when an arbitrary value is given to the first decision variable vector. In conventional genetic algorithms (GAs) for the problem, an individual is often expressed by the value of the first decision variable vector. In evaluating the individual, the value of the remaining decision variable vector is determined by metaheuristics or greedy algorithms. However, such GAs are time-consuming or not general-purpose. We propose a GA with a neural network model to estimate the optimal objective function values of the subproblems. Experimental results compared to other GAs show that the proposed method is effective.

show abstract

Solving combinatorial bi-level optimization problems using multiple populations and migration schemes

Cited by 9 publications

References 54 publications

Integration of production scheduling and energy-cost optimization using Mean Value Cross Decomposition

Integration of production scheduling and energy-cost optimization using Mean Value Cross Decomposition

Discretization-Based Feature Selection as a Bilevel Optimization Problem

Genetic Algorithm with Machine Learning to Estimate the Optimal Objective Function Values of Subproblems

Contact Info

Product

Resources

About