dc.contributor.author |
Dymond, Antoine Smith Dryden
|
|
dc.contributor.author |
Kok, Schalk
|
|
dc.contributor.author |
Heyns, P.S. (Philippus Stephanus)
|
|
dc.date.accessioned |
2017-04-24T07:59:54Z |
|
dc.date.issued |
2017-03 |
|
dc.description.abstract |
Control parameter studies assist practitioners to select optimization algorithm parameter values which are appropriate
for the problem at hand. Parameters values are well-suited to a problem if they result in a search which is
effective given that problem’s objective function(s), constraints and termination criteria. Given these considerations
a many objective tuning algorithm named MOTA is presented. MOTA is specialized for tuning a stochastic optimization
algorithm according to multiple performance measures each over a range of objective function evaluation
budgets. MOTA’s specialization consist of four aspects; 1) a tuning problem formulation which consists of both a
speed objective and a speed decision variable, 2) a control parameter tuple assessment procedure which utilizes information
from a single assessment run’s history to gauge that tuple’s performance at multiple evaluation budgets,
3) a preemptively terminating resampling strategy for handling the noise present when tuning stochastic algorithms,
and 4) the use of bi-objective decomposition to assist in many objective optimization. MOTA combines these aspects
together with DE operators to search for effective control parameter values. Numerical experiments which consisted
of tuning NSGA-II and MOEA/D demonstrate that MOTA is effective at many objective tuning. |
en_ZA |
dc.description.department |
Mechanical and Aeronautical Engineering |
en_ZA |
dc.description.embargo |
2017-06-30 |
|
dc.description.librarian |
hb2017 |
en_ZA |
dc.description.sponsorship |
The National Research Foundation (NRF) of South
Africa. |
en_ZA |
dc.description.uri |
http://www.mitpressjournals.orgloi/evco |
en_ZA |
dc.identifier.citation |
Dymond, AS, Kok, S & Heyns, PS 2017, 'MOTA : a many-objective tuning algorithm specialized for tuning undermultiple objective function evaluation budgets', Evolutionary Computation, vol. 25, no. 1, pp. 113-141. |
en_ZA |
dc.identifier.issn |
1063-6560 (print) |
|
dc.identifier.issn |
1530-9304 (online) |
|
dc.identifier.other |
10.1162/EVCO_a_00163 |
|
dc.identifier.uri |
http://hdl.handle.net/2263/60014 |
|
dc.language.iso |
en |
en_ZA |
dc.publisher |
Massachusetts Institute of Technology Press |
en_ZA |
dc.rights |
© 2017 The MIT Press |
en_ZA |
dc.subject |
Objective function evaluation budgets |
en_ZA |
dc.subject |
Tuning |
en_ZA |
dc.subject |
Many-objective optimization |
en_ZA |
dc.title |
MOTA : a many-objective tuning algorithm specialized for tuning undermultiple objective function evaluation budgets |
en_ZA |
dc.type |
Article |
en_ZA |