Runtime analysis of different Approaches to select conflicting auxiliary objectives in the generalized OneMax problem

Allbwn ymchwil: Pennod mewn Llyfr/Adroddiad/Trafodion CynhadleddTrafodion Cynhadledd (Nid-Cyfnodolyn fathau)

Crynodeb

It has been shown that single-objective optimization may be improved by introducing auxiliary objectives. Ideally, auxiliary objectives should be designed to be independent. However, in practice, this is not always easy or possible and the objectives may be conflicting. Particularly, this happens in the Job-Shop Scheduling problem and in certain search-based software engineering problems. We theoretically analyse different approaches of using auxiliary objectives on the general ONEMAX problem with conflicting auxiliary objectives ONEMAX and ZEROMAX. In most of the considered methods, the optimized objectives are selected dynamically. We also consider a method, where the same objectives are optimized during the whole run. We show that dynamic selection of conflicting auxiliary objectives combined with explicit optimization of the target objective has the best runtime of O(nlogn) among all the considered methods.

Iaith wreiddiolSaesneg
Teitl2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016
CyhoeddwrIEEE Press
Nifer y tudalennau7
ISBN (Electronig)9781509042401
Dynodwyr Gwrthrych Digidol (DOIs)
StatwsCyhoeddwyd - 09 Chwef 2017
Cyhoeddwyd yn allanolIe
Digwyddiad2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016 - Athens, Groeg
Hyd: 06 Rhag 201609 Rhag 2016

Cyfres gyhoeddiadau

Enw2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016

Cynhadledd

Cynhadledd2016 IEEE Symposium Series on Computational Intelligence, SSCI 2016
Gwlad/TiriogaethGroeg
DinasAthens
Cyfnod06 Rhag 201609 Rhag 2016

Ôl bys

Gweld gwybodaeth am bynciau ymchwil 'Runtime analysis of different Approaches to select conflicting auxiliary objectives in the generalized OneMax problem'. Gyda’i gilydd, maen nhw’n ffurfio ôl bys unigryw.

Dyfynnu hyn