ADAPTABILITY ANALYSIS OF APPARENT FRONT RANKING

Mihai NEGHINĂ

Abstract


In the context of multi-objective optimization through genetic algorithms, the ranking of individuals is a delicate and consequential operation. Introduced in 2020, the method of Apparent Front Ranking has shown potential, being successfully included in a GA. The current paper offers a deeper analysis of the flexibility of AFR, with insights about the evolution of ranks as the AF template power is modified, showing excellent Kendall and Pearson correlations with the reference hierarchical methods in uniform datasets of up to 10 dimensions. By adjusting the AF template power, the best Kendall rank correlation is above 0.79 with all reference methods, while the best Pearson correlation is above 0.95 with Saaty’s AHP and above 0.75 with Köppen’s average FPD.

Full Text:

PDF

References


Katoch S. et al., A review on genetic algorithms: past, present, and future. Multimedia Tools and Applications 80(5), 8091–8126, 2021. DOI

Pareto V., Cours d’economique politique. Vol. I/II., F. Rouge, 1896-7.

Zitzler E., Evolutionary Algorithms for Multiobjective Optimization: Methods and Applications. Swiss Federal Institute of Technology Zurich, Switzerland, 1999.

ISBN-13:‎ 9783826568312

Deb K., Multi-objective optimization using evolutionary algorithms. Wiley, UK, 2001.

ISBN-13: 9780471873396

Deb K., Saxena D., On finding Pareto-optimal solutions through dimensionality reduction for certain large-dimensional multi-objective optimization problems. Kangal report 2005011, Indian Institute of Technology, 2005. SemanticsScholarLink

Laumanns M., Thiele L., Deb K., Zitzler E. Combining convergence and diversity in evolutionary multiobjective optimization. Evol. Comp. 10.3, 263–282, 2002. DOI

Adra S., Fleming P. Diversity management in evolutionary many-objective optim. Evol. Comp. 15.2, 183–195, 2011. DOI

Jiang S., Yang S., Convergence versus diversity in multiobjective optimization, PPSN XIV 9921, pp. 984–993, 2016. DOI

Rostami S., Neri F., A Fast Hypervolume Driven Selection Mechanism for Many-Objective Optimisation Problems. Swarm and Evol. Comp. 34, 50–67, 2017. DOI

Liu Z.Z. et al., A Many-Objective Evolutionary Algorithm with Angle-Based Selection and Shift-Based Density Estimation. Information Sciences 509, 400-419, 2020. SemanticsScholarLink

Farina M., Amato P. Fuzzy optimality and evolutionary multiobjective optimization. Evolutionary Multi-Criterion Optimization EMO2003, 58–72, 2003. DOI

Köppen M. et al. Fuzzy-Pareto-Dominance and its Application in Evolutionary Multi-objective Optimization. Evolutionary Multi-Criterion Optimization, LN in Comp. Science 3410, 399–412, 2005. DOI

Silva R.C., Yamakami A. Definition of Fuzzy Pareto-Optimality by Using Possibility Theory. IFSA/EUSFLAT, 1234–1239, 2009. SemanticsScholarLink

Farina M., Amato P. On the optimal solution definition for many-criteria optimization problems. NAFIPS-FLINT, 233–238, 2002. DOI

Neghină M., Vinţan L., Apparent Front Ranking: novel population ranking method for genetic multi-objective algorithms. Proc. of the Romanian Academy A, 21 (2), 179–186, 2020. AcadRoLink

Neghină M. et al., A competitive new multi-objective optimization genetic algorithm based on Apparent Front Ranking, Engineering Applications of Artificial Intelligence, Vol. 132(107870), 2024. DOI

Pătrăușanu A. et al., A Systematic Review of Multi-Objective Evolutionary Algorithms Optimization Frameworks Processes 12(5):869, 2024. DOI

Saaty T.L. The Analytical Hierarchy Process. McGraw-Hill, New York, 1980.

ISBN-13: 9780070543713

Kendall M.G.. Rank Correlation Methods. Hafner Publishing Co., New York, 1955.

ISBN-13: 9780852641996

Pearson K. Note on regression and inheritance in the case of two parents. Proc. Royal Soc. London 58, 1895. DOI

Fitzgibbon A. et al., Direct least square fitting of ellipses. Pattern and Analysis Machine Intel. 21.5, 476–480, 1999. DOI

Deb K. et al., A fast and elitist multiobjective genetic algorithm: NSGA-II. Evol. Comp. 6.2, 182–197, 2002. DOI

Zitzler E. et al., SPEA2: Improving the strength pareto evolutionary algorithm for multiobjective optimization., EUROGEN, 95–100, 2001. DOI


Refbacks

  • There are currently no refbacks.


JOURNAL INDEXED IN :