Towards objective measures of algorithm performance across instance space

Smith-Miles, Kate, Baatar, Davaatseren, Wreford, Brendan and Lewis, Rhyd ORCID: https://orcid.org/0000-0003-1046-811X 2014. Towards objective measures of algorithm performance across instance space. Computers & Operations Research 45 , pp. 12-24. 10.1016/j.cor.2013.11.015

Preview

PDF - Submitted Pre-Print Version Download (2MB) | Preview

Abstract

This paper tackles the difficult but important task of objective algorithm performance assessment for optimization. Rather than reporting average performance of algorithms across a set of chosen instances, which may bias conclusions, we propose a methodology to enable the strengths and weaknesses of different optimization algorithms to be compared across a broader instance space. The results reported in a recent Computers and Operations Research paper comparing the performance of graph coloring heuristics are revisited with this new methodology to demonstrate (i) how pockets of the instance space can be found where algorithm performance varies significantly from the average performance of an algorithm; (ii) how the properties of the instances can be used to predict algorithm performance on previously unseen instances with high accuracy; and (iii) how the relative strengths and weaknesses of each algorithm can be visualized and measured objectively.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Mathematics
Subjects:	Q Science > QA Mathematics
Uncontrolled Keywords:	Comparative analysis; Heuristics; Graph coloring; Algorithm selection; Performance prediction
Additional Information:	Pdf uploaded in accordance with the publisher’s policy at http://www.sherpa.ac.uk/romeo/issn/0305-0548/(accessed 07/08/2014)
Publisher:	Elsevier
ISSN:	0305-0548
Last Modified:	13 Apr 2024 08:15
URI:	https://orca.cardiff.ac.uk/id/eprint/53966

Citation Data

Cited 106 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item