Decomposition-based multiobjective optimization with bicriteria assisted adaptive operator selection

Lin, Wu; Lin, Qiuzhen; Ji, Junkai; Zhu, Zexuan; Coello, Carlos A. Coello; Wong, Ka-Chun

doi:10.1016/j.swevo.2020.100790

Cited by 16 publications

(6 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other deterministic alternatives to adapt to fast rewards dynamic are by using bandit based operator selector [26]. A recent AOS strategy based on the current solution status is called bicriteria assisted AOS (BAOS) [27]. By using the current solution state instead of the operator performance, BAOS does not need credit assignment, and therefore straightforward to be adopted into any MOEAs.…”

Section: Minimize 𝐹(𝑥 ⃗)mentioning

confidence: 99%

“…One variant employs a tournament parent selection mechanism similar to NSGA-II but with HAD as the crowdedness measure, denoted as MO-ALFPAT. The other variant, MO-ALFPAB, with bicriteria adaptive operator selection (BAOS) [27] to choose the appropriate operator based on the current status of the solution. Both variants improve the performance of MO-ALFPA, especially MO-ALFPAT.…”

Section: Minimize 𝐹(𝑥 ⃗)mentioning

confidence: 99%

“…else Generate 𝑥 𝑖,𝑗 𝑡+1 , ∀ 𝑗 ∈ [1,4] with Eq. ( 3)-( 6) (12) Lin et al [27] argued that HAD can reflect the crowdedness of each solution in the local search space. In addition, it is also shown in [29] that crowding distance in NSGA-II is sensitive to outlier and therefore may not accurately reflect the actual crowdedness of solutions in the population.…”

Section: Mo-alfpamentioning

confidence: 99%

“…( 3)-( 6) The second variant, MO-ALFPAB, select an appropriate operator to explore a solution instead of selecting the appropriate solution to explore as in MO-ALFPA. An adaptive operator selection called bi-criteria adaptive operator selection (BAOS) proposed by Lin et al [27] is used in MO-ALFPAB.…”

Section: Mo-alfpamentioning

confidence: 99%

See 3 more Smart Citations

Adaptive Lévy Flower Pollination Algorithm for Multiple Objective Optimization

2022

IJIES

View full text Add to dashboard Cite

The adaptive Lévy flower pollination algorithm (ALFPA) is a recent addition to variants of flower pollination algorithm (FPA). Despite its excellent performance on single objective problem instances, it has shown inefficiency in the number of function evaluation (FE). Inspired by this, this paper proposed two algorithms extending ALFPA to solve multiple objective problem while also improving FE efficiency. The first algorithm proposed is ALFPA with non-dominated sorting denoted as MO-ALFPA alongside two variants which are proposed to improve its FE efficiency denoted as MO-ALFPAT and MO-ALFPAB. The second proposed algorithm, MOEA/D-ALFPA, uses decomposition strategy instead of non-dominated sorting on MO-ALFPA. The empirical study on two benchmark suits shows that MO-ALFPAT and MOEA/D-ALFPA performed better than other methods. Furthermore, MO-ALFPAT and MOEA/D-ALFPA produced the best results in three benchmark instances, based on inverted generational distance indicator, and two and three best results based on the hypervolume indicator, respectively.

show abstract

Section: Minimize 𝐹(𝑥 ⃗)mentioning

confidence: 99%

Section: Minimize 𝐹(𝑥 ⃗)mentioning

confidence: 99%

Section: Mo-alfpamentioning

confidence: 99%

Section: Mo-alfpamentioning

confidence: 99%

See 2 more Smart Citations

Adaptive Lévy Flower Pollination Algorithm for Multiple Objective Optimization

2022

IJIES

View full text Add to dashboard Cite

show abstract

“…Over the several decades, multi-objective evolutionary algorithms (MOEAs) have been well studied and shown advantages for solving MOPs. Most existing MOEAs can be broadly grouped into three categories, including Pareto-based MOEAs [9][10][11][12][13], decomposition-based MOEAs [14][15][16][17][18] and indicator-based MOEAs [19][20][21][22][23].…”

Section: Introductionmentioning

confidence: 99%