On the integrity of performance comparison for evolutionary multi-objective optimisation algorithms

Authors: Wilson, K. and Rostami, S.

Journal: Advances in Intelligent Systems and Computing

Volume: 840

Pages: 3-15

ISBN: 9783319979816

ISSN: 2194-5357

DOI: 10.1007/978-3-319-97982-3_1

Abstract:

This paper proposes the notion that the experimental results and performance analyses of newly developed algorithms in the field of multi-objective optimisation may not offer sufficient integrity for hypothesis testing. The reason for this is that many implementations exist of the same optimisation algorithms, and these may vary in behaviour due to the interpretation of the developer. This is demonstrated through the comparison of three implementations of the popular Non-dominated Sorting Genetic Algorithm II (NSGA-II) from well-regarded frameworks using the hypervolume indicator. The results show that of the thirty considered comparison cases, only four indicate that there was no significant difference between the performance of either implementation.

https://eprints.bournemouth.ac.uk/31060/

Source: Scopus

On the Integrity of Performance Comparison for Evolutionary Multi-objective Optimisation Algorithms

Authors: Wilson, K. and Rostami, S.

Journal: ADVANCES IN COMPUTATIONAL INTELLIGENCE SYSTEMS (UKCI)

Volume: 840

Pages: 3-15

eISSN: 2194-5365

ISBN: 978-3-319-97981-6

ISSN: 2194-5357

DOI: 10.1007/978-3-319-97982-3_1

https://eprints.bournemouth.ac.uk/31060/

Source: Web of Science (Lite)

On the Integrity of Performance Comparison for Evolutionary Multi-objective Optimisation Algorithms

Authors: Wilson, K. and Rostami, S.

Conference: UKCI 2018 : 18TH ANNUAL UK WORKSHOP ON COMPUTATIONAL INTELLIGENCE

Dates: 5-7 September 2018

https://eprints.bournemouth.ac.uk/31060/

Source: Manual

On the Integrity of Performance Comparison for Evolutionary Multi-objective Optimisation Algorithms

Authors: Wilson, K. and Rostami, S.

Conference: UKCI 2018: 18th Annual UK Workshop on Computational Intelligence

Abstract:

This paper proposes the notion that the experimental results and performance analyses of newly developed algorithms in the field of multi-objective optimisation may not offer sufficient integrity for hypothesis testing. This is demonstrated through the multiple comparison of three implementations of the popular Non-dominated Sorting Genetic Algorithm II (NSGA-II) from well-regarded frameworks using the hypervolume indicator. The results show that of the thirty considered comparison cases, only four indicate that there was no significant difference between the performance of either implementation.

https://eprints.bournemouth.ac.uk/31060/

http://ukci2018.uk/

Source: BURO EPrints