Experimental Results for the study "The Hypervolume Newton Method for Constrained Multi-Objective Optimization Problems"

This repository contains the experiment results (raw data in NPZ and CSV format and Latex tables) for the study "The Hypervolume Newton Method for Constrained Multi-Objective Optimization Problems", which is accepted in Mathematical and Computational Applications journal.

The preprint version of the related paper is already online: 

Wang, H.; Emmerich, M.; Deutz, A.; Hernández, V.A.S.; Schütze, O. The Hypervolume Newton Method for Constrained Multi-objective Optimization Problems. Preprints 2022, 2022110103 (doi: 10.20944/preprints202211.0103.v1).

Data description: we benchmarked three algorithms: (1) the standalone Hypervolume Netwon Method (HVN), (2) NSGA-III, and (3) the hybridization of the standalone HVN and NSGA-III on several artificial problems.

For the standalone HVN algorithm, we tested it on three simple artificial test problems - P1, P2, and P3 (proposed in the above paper):

2D-example-50*.tex: problem P1

3D-example1*.tex: problem P2

3D-example2*.tex: problem P3

For NSGA-III and the hybridization, we tested them on the equality-constrained DTLZ and Inverted DTLZ (IDTLZ) problems:

Eq1DTLZ.*npz: DTLZ problems

Eq1IDTLZ*.npz: IDTLZ problems