Deep Statistical Comparison for Meta-heuristic Stochastic Optimization Algorithms by Tome Eftimov & Peter Korošec

Author:Tome Eftimov & Peter Korošec
Language: eng
Format: epub
ISBN: 9783030969172
Publisher: Springer International Publishing

9.864

2.439

9.862

2.509

0.971

9.693

6.259

0.387

6.309

1.579

3.848

3.970

6.208

0.396

9.908

1.510

0.579

6.307

0.250

0.397

24.844

6.308

0.966

15.842

2.381

0.056

6.305

5.576

1.577

3.882

0.957

0.604

23.460

2.244

0.391

6.121

2.490

6.271

14.341

1.584

0.244

3.930

2.497

2.483

9.937

To avoid the dependence on the order of the independent runs, the same combination of the three algorithms was analyzed using the set of 22 benchmark problems by applying the Monte Carlo pDSC ranking scheme for the same range of the practical thresholds presented in Table 6.7. Table 6.10 presents the p-values obtained using the Monte Carlo pDSC ranking scheme. Comparing them with the p-values obtained using the sequential pDSC ranking scheme, there are differences that happen for some parasitical thresholds: , , and . This result indicates that for these practical thresholds the order of the independent runs affects the ranking process.Table 6.10Statistical comparison of three algorithms using the Monte Carlo variant of the pDSC

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.