An evolutionary many-objective algorithm based on decomposition and hierarchical clustering selection (original) (raw)

Abstract

In recent years, many multi-objective evolutionary algorithms have been proposed to solve many-objective optimization problems with regular Pareto front. These algorithms have shown good performance in balancing convergence and diversity. However, in the high-dimensional objective space, the non-dominated solutions increases exponentially as the number of objectives increases. The metrics to evaluate algorithm performance are also computationally intensive. In particular, solving the many-objective optimization problem of the irregular Pareto front faces great challenges. Moreover, many-objective evolutionary algorithms, do not easily show their convergence and diversity through visualization, as multi-objective evolutionary algorithms do. To address these problems, a many-objective optimization algorithm based on decomposition and hierarchical clustering selection is proposed in this paper. First, a set of uniformly distributed reference vectors divides non-dominanted individuals into different sub-populations, and then candidate solutions are selected based on the aggregation function values in the sub-populations. Second, a set of adaptive reference vectors is used to rank the dominant individuals in the population and retain promising candidate solutions. Third, a hierarchical clustering selection strategy is used to enable solutions with good convergence to be selected. Finally, a diversity maintenance strategy is used to remove solutions with poor diversity. The experimental results show that the proposed algorithm EA-DAH has advantages over other comparative algorithms in many-objective optimization problems with irregular Pareto fronts.

Access this article

Log in via an institution

Subscribe and save

• Starting from 10 chapters or articles per month
• Access and download chapters and articles from more than 300k books and 2,500 journals
• Cancel anytime View plans

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Notes

References

1. Lygoe RJ, Cary M, Fleming PJ (2013) A real-world application of a many-objective optimisation complexity reduction process. International Conference on Evolutionary Multi-Criterion Optimization
2. Reed PM, David H (2014) Evolving many-objective water management to exploit exascale computing. Water Resour Res 50(10):8367–8373
   Article Google Scholar
3. MolinaLopez JM, Herrero JG, Berlanga A (2009) Effective evolutionary algorithms for many-specifications attainment: Application to air traffic control tracking filters. IEEE Trans Evol Comput 13(1):151–168
   Article Google Scholar
4. Fleming PF, Purshouse RC, Lygoe RJ (2005) Many-objective optimization: An engineering design perspective 3410:14–32
5. Yang S, Li M, Liu X, Zheng J (2013) A grid-based evolutionary algorithm for many-objective optimization. IEEE Trans Evol Comput 17(5):721–736
   Article Google Scholar
6. Ye TXZ, Jin Y (2014) Shift-based density estimation for pareto-based algorithms in many-objective optimization. IEEE Trans Evol Comput 18(3):348–365
   Article Google Scholar
7. Zhang X, Ye T, Jin Y (2015) A knee point driven evolutionary algorithm for many-objective optimization. IEEE Trans Evol Comput 19(6):761–776
   Article Google Scholar
8. Yi X, Zhou Y, Li M, Chen Z (2017) A vector angle-based evolutionary algorithm for unconstrained many-objective optimization. IEEE Trans Evol Comput 21(1):131–152
   Article Google Scholar
9. Bader J, Zitzler E (2011) Hype: An algorithm for fast hypervolume-based many-objective optimization. IEEE Trans Evol Comput 19(1):45–76
   Google Scholar
10. Gmez RH, Coello CA (2015) Coello Improved metaheuristic based on the r2 indicator for many-objective optimization, pp 679C686
11. Zhang Q, Li H (2008) Moea/d: a multiobjective evolutionary algorithm based on decomposition. IEEE Trans Evol Comput 11(6):712–731
    Article Google Scholar
12. Li K, Deb K, Zhang Q, Kwong S (2015) An evolutionary many-objective optimization algorithm based on dominance and decomposition. IEEE Trans Evol Comput 19(5):694–716
    Article Google Scholar
13. Cheng R, Jin Y, Olhofer M, Sendhoff B (2016) A reference vector guided evolutionary algorithm for many-objective optimization. IEEE Trans Evol Comput 20(5):773–791
    Article Google Scholar
14. Deb K, Jain H (2014) An evolutionary many-objective optimization algorithm using reference-point-based nondominated sorting approach, part I: Solving problems with box constraints. IEEE Trans Evol Comput 18(4):577–601
    Article Google Scholar
15. Jiang S, Yang S (2017) A strength pareto evolutionary algorithm based on reference direction for multi-objective and many-objective optimization. IEEE Trans Evol Comput 21(3):329– 346
    Article Google Scholar
16. Ishibuchi H, Tsukamoto N, Nojima Y (2008) Evolutionary many-objective optimization A short review. In 2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence), pp 2419–2426
17. Cai X, Yang Z, Fan Z, Zhang Q (2017) Decomposition-based-sorting and angle-based-selection for evolutionary multiobjective and many-objective optimization. IEEE Trans Cybern 47(9):2824–2837
    Article Google Scholar
18. Xu H, Zeng W, Zhang D, Xiangxiang Z (2019) Moea/hd: a multiobjective evolutionary algorithm based on hierarchical decomposition. IEEE Trans Cybern 49(2):517–526
    Article Google Scholar
19. Denysiuk R, Costa L, Santo IE (2014) Clustering-based selection for evolutionary many-objective optimization. In International Conference on Parallel Problem Solving from Nature
20. Deb K (2001) Multiobjective optimization using evolutionary algorithms. Wiley, New York
    MATH Google Scholar
21. Zhou A, Jin Y, Zhang Q, Sendhoff B, Tsang E (2006) Combining model-based and genetics-based offspring generation for multi-objective optimization using a convergence criterion. In: 2006 IEEE International Conference on Evolutionary Computation, pp 892–899
22. Li K, Kwong S, Cao J, Li M, Zheng J, Shen R (2012) Achieving balance between proximity and diversity in multi-objective evolutionary algorithm. Inf Sci 182(1):220–242
    Article MathSciNet Google Scholar
23. Ye T, Cheng R, Zhang X, Cheng F, Jin Y (2018) An indicator based multi-objective evolutionary algorithm with reference point adaptation for better versatility. IEEE Trans Evol Comput 22(4):609–622
    Article Google Scholar
24. Wu M, Li K, Kwong S, Zhang Q (2020) Evolutionary many-objective optimization based on adversarial decomposition. IEEE Trans Cybern 50(2):753–764
    Article Google Scholar
25. Sam L, Zhang KQ, Mengyuan W, Zhang J (2019) Learning to decompose: A paradigm forn decomposition-based multiobjective optimization. IEEE Trans Evol Comput 23(3):376–390
    Article Google Scholar
26. Das I, Dennis JE (1996) Normal-boundary intersection: a new method for generating the pareto surface in nonlinear multicriteria optimization problems. Siam J Optim 8(3):631–657
    Article MathSciNet Google Scholar
27. Cheng R, Jin Y, Narukawa K, Sendhoff B (2015) A multiobjective evolutionary algorithm using gaussian processbased inverse modeling. IEEE Trans Evol Comput 19(6):838–856
    Article Google Scholar
28. Dickinson AW (2002) Experiments with mixtures: Designs, models, and the analysis of mixture data. Technometrics 24(2):161–162
    Article Google Scholar
29. Qi Y, Ma X, Liu F, Jiao L, Sun J, Wu J (2014) Moea/d with adaptive weight adjustment. IEEE Trans Evol Comput 22(2):231–264
    Google Scholar
30. Agrawal RB, Deb K, Agrawal RB (1994) Simulated binary crossover for continuous search space. Compl Syst 9(3):115–148
    MathSciNet MATH Google Scholar
31. Deb K, Goyal M (1996) A combined genetic adaptive search (geneas) for engineering design. Comput Sci Inf 26(4):30–45
    Google Scholar
32. Deb K, Thiele L, Laumanns M, Zitzler E (2001) Scalable test problems for evolutionary multiobjective optimization, pp 105–145
33. Jain H, Deb K (2019) An evolutionary many-objective optimization algorithm using reference point based nondominated sorting approach, part ii: Handling constraints and extending to an adaptive approach. IEEE Trans Evol Comput 18(4):602–622
    Article Google Scholar
34. Deb K (2005) Scalable test problems for evolutionary multiobejctive optimization. Evolutionary Multiobjective Optimization Theoretical Advances & Applications
35. Tian XZY, Cheng R, Jin Y (2017) Platemo: A matlab platform for evolutionary multiobjective optimization [educational forum]. IEEE Comput Intell Mag 12(4):73–87
    Article Google Scholar
36. Coello CA, Nareli Cruz Corts. (2005) Solving multiobjective optimization problems using an artificial immune system. Coello Genet Programm Evolvable Mach 6(2):163–190
    Article Google Scholar
37. Tian X, Miqing Z, Ye L, Cheng R, Jin Y (2019) Diversity assessment of multi-objective evolutionary algorithms Performance metric and benchmark problems [research frontier]. IEEE Comput Intell Mag 14(3):61–74
    Article Google Scholar
38. He X, Zhou Y, Chen Z, Zhang Q (2019) Evolutionary many-objective optimization based on dynamical decomposition. IEEE Trans Evol Comput 23(3):361–375
    Article Google Scholar
39. Yuan Y, Xu H, Bo W, Bo Z, Yao X (2016) Balancing convergence and diversity in decomposition-based many-objective optimizers. IEEE Trans Evol Comput 20(2):180–198
    Article Google Scholar
40. Durillo JJ, Nebro AJ (2011) jmetal:a java framework for multi-objective optimization. Adv Eng Softw 42(10):760–771
    Article Google Scholar
41. Ishibuchi HMH, Yu S, Nojima Y (2017) Performance of decomposition-based many-objective algorithms strongly depends on pareto front shapes. IEEE Trans Evol Comput 21(2):169–190
    Article Google Scholar
42. Phillips P, Yang J, Sun P, Shuihua Y, Zhang W (2016) Magnetic resonance brain classification by a novel binary particle swarm optimization with mutation and time-varying acceleration coefficients. Biomedizinische Technik Biomedical engineering
43. Li J, Wu X, Wang H, Liu B, Dong Z, Wang S, Yang M, Zhang Y (2017) Texture analysis method based on fractional fourier entropy and fitness-scaling adaptive genetic algorithm for detecting left-sided and right-sided sensorineural hearing loss. Fund Inf 151(1-4):505–521
    MathSciNet Google Scholar
44. Li J, Wu X, Wang H, Liu B, Dong Z, Wang S, Yang M, Zhang Y (2017) Investigating the effect of imbalance between convergence and diversity in evolutionary multi-objective algorithms. IEEE Trans Evol Comput 21(3):408–425
    Google Scholar

Download references

Acknowledgements

This research is partly supported by the Natural Science Foundation of China (Grant No. 11871279 and 61971234), Humanity and Social Science Youth foundation of Ministry of Education of China (Grant No. 12YJCZH179), the Natural Science Foundation of the Jiangsu Higher Education Institutions of China (Grant No. 16KJA110001). Thanks all authors for providing the source codes of the comparison algorithms.

Author information

Authors and Affiliations

1. School of Mathematical Sciences, Nanjing Normal University, Nanjing, 210023, People’s Republic of China
   Yuehong Sun, Kelian Xiao, Siqiong Wang & Qiuyue Lv
2. Jiangsu Key Laboratory for Numerical Simulation of Large Scale Complex Systems, Nanjing, 210023, People’s Republic of China
   Yuehong Sun

Authors

1. Yuehong Sun
2. Kelian Xiao
3. Siqiong Wang
4. Qiuyue Lv

Corresponding author

Correspondence toYuehong Sun.

Ethics declarations

Conflict of Interest

The authors declare that they have no conflict of interest.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

Fig. 12

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ1 with 3-objective

Fig. 13

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ1 with 4-objective

Fig. 14

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ1 with 5-objective

Fig. 15

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ1 with 6-objective

Fig. 16

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ1 with 8-objective

Fig. 17

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ1 with 10-objective

Fig. 18

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ2 with 3-objective

Fig. 19

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ2 with 4-objective

Fig. 20

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ2 with 5-objective

Fig. 21

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ2 with 6-objective

Fig. 22

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ2 with 8-objective

Fig. 23

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ2 with 10-objective

Fig. 24

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ3 with 3-objective

Fig. 25

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ3 with 4-objective

Fig. 26

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ3 with 5-objective

Fig. 27

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ3 with 6-objective

Fig. 28

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ3 with 8-objective

Fig. 29

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ3 with 10-objective

Fig. 30

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ4 with 3-objective

Fig. 31

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ4 with 4-objective

Fig. 32

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ4 with 5-objective

Fig. 33

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ4 with 6-objective

Fig. 34

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ4 with 8-objective

Fig. 35

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ4 with 10-objective

Fig. 36

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ5 with 3-objective

Fig. 37

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ5 with 4-objective

Fig. 38

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ5 with 5-objective

Fig. 39

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ5 with 6-objective

Fig. 40

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ5 with 8-objective

Fig. 41

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ5 with 10-objective

Fig. 42

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ6 with 3-objective

Fig. 43

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ6 with 4-objective

Fig. 44

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ6 with 5-objective

Fig. 45

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ6 with 6-objective

Fig. 46

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ6 with 8-objective

Fig. 47

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ6 with 10-objective

Fig. 48

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ7 with 3-objective

Fig. 49

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ7 with 4-objective

Fig. 50

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ7 with 5-objective

Fig. 51

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ7 with 6-objectiv

Fig. 52

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ7 with 8-objective

Fig. 53

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on DTLZ7 with 10-objective

Fig. 54

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C1DTLZ1 with 3-objective

Fig. 55

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C1DTLZ1 with 4-objective

Fig. 56

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C1DTLZ1 with 5-objective

Fig. 57

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C1DTLZ1 with 6-objective

Fig. 58

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C1DTLZ1 with 8-objective

Fig. 59

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C1DTLZ1 with 10-objective

Fig. 60

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C2DTLZ2 with 3-objective

Fig. 61

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C2DTLZ2 with 4-objective

Fig. 62

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C2DTLZ2 with 5-objective

Fig. 63

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C2DTLZ2 with 6-objective

Fig. 64

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C2DTLZ2 with 8-objective

Fig. 65

Parallel Coordinate Plots for True PFs and PF Approximations Obtained by EA-DAH, DDEANS, RVEA, NSGA-III, VaEA, EFRRR, and MOEA/DD on C2DTLZ2 with 10-objective

Rights and permissions

About this article

Cite this article

Sun, Y., Xiao, K., Wang, S. et al. An evolutionary many-objective algorithm based on decomposition and hierarchical clustering selection.Appl Intell 52, 8464–8509 (2022). https://doi.org/10.1007/s10489-021-02669-9

Download citation

• Accepted: 18 May 2021
• Published: 30 October 2021
• Version of record: 30 October 2021
• Issue date: June 2022
• DOI: https://doi.org/10.1007/s10489-021-02669-9

Keywords