Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/23150
Title: | Analyzing the Impact of the Adaptive Clearing Mechanism on Algorithm Accuracy in Variable Mesh Optimization | Authors: | VANHOENSHOVEN, Frank NAPOLES RUIZ, Gonzalo CREEMERS, Mathijs VANHOOF, Koen Espinosa, Maikel Leon |
Issue Date: | 2016 | Source: | The 2016 IEEE Symposium Series on Computational Intelligence (IEEE SSCI 2016), Athens, Greece, 6-9/12/2016 | Abstract: | The area of population-based meta-heuristics has been researched extensively in recent years. The focus of this research has been on finding improvements and variations to existing algorithms while the inner details, that are treated as a black box, remain poorly understood. The purpose of this paper is to uncover the detailed behavior of Variable Mesh Optimization (VMO), a population-based meta-heuristic, and describe the patterns that drive the algorithm in finding new optima. Our results suggest that, in VMO, the improvement of the best solution is strongly correlated with its adaptive clearing mechanism. It is observed that each relaxation of the threshold that is used by the mechanism, is likely to increase the accuracy of the final solution. These findings suggest that future research, aiming to improve algorithm accuracy, could focus on improving the adaptive clearing mechanism in order to increase the likelihood of creating superior algorithms. | Keywords: | vmo; pmh; variable mesh optimization; population-based heuristics; adaptive clearing mechanism; optimization; algorithm | Document URI: | http://hdl.handle.net/1942/23150 | Category: | C2 | Type: | Conference Material |
Appears in Collections: | Research publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Analyzing the Impact of the Adaptive Clearing Mechanism on Algorithm Accuracy.pdf | Conference material | 378.06 kB | Adobe PDF | View/Open |
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.