Persistent URL of this record https://hdl.handle.net/1887/123184
Documents
-
- Download
- Title pages_Contents
-
open access
-
- Download
- Chapter 4
-
open access
- Full text at publishers site
-
- Download
- Chapter 5
-
open access
- Full text at publishers site
-
- Download
- Chapter 6
-
open access
- Full text at publishers site
-
- Download
- Appendix_Bibliography_Glossary
-
open access
-
- Download
- Summary in Dutch
-
open access
-
- Download
- Curriculum Vitae
-
open access
-
- Download
- Propositions
-
open access
In Collections
This item can be found in the following collections:
Optimally weighted ensembles of surrogate models for sequential parameter optimization
In surrogate-model-assisted optimization, selecting the right modeling technique without preliminary knowledge about the objective function can be challenging. It might be beneficial if the algorithm trains many different surrogate models and selects the model with the smallest training error. This approach is known as model selection.
In this thesis, a generalization of this approach is developed. Instead of choosing a single model, the optimal convex combinations of model predictions is used to combine surrogate models into one more accurate ensemble surrogate model.
This approach is studied in a fundamental way, by first evaluating minimalistic ensembles of only two surrogate models in detail and then proceeding to ensembles with more surrogate models.
Show moreIt is a common technique in global optimization with expensive black-box functions to learn a surrogate-model of the response function from past evaluations and use it to decide on the location of future evaluations.
In surrogate-model-assisted optimization, selecting the right modeling technique without preliminary knowledge about the objective function can be challenging. It might be beneficial if the algorithm trains many different surrogate models and selects the model with the smallest training error. This approach is known as model selection.
In this thesis, a generalization of this approach is developed. Instead of choosing a single model, the optimal convex combinations of model predictions is used to combine surrogate models into one more accurate ensemble surrogate model.
This approach is studied in a fundamental way, by first evaluating minimalistic ensembles of only two surrogate models in detail and then proceeding to ensembles with more surrogate models.
Finally, the approach is adopted and evaluated in the context of sequential parameter optimization. Besides discussing the general strategy, the optimal frequency of learning the convex combination of weights is investigated.
The results provide insights into the performance, scalability, and robustness of the approach.
Show less
- All authors
- Echtenbruck, M.M.
- Supervisor
- Bäck, T.W.H.; Bartz-Beielstein, T.
- Co-supervisor
- Emmerich, M.T.M.
- Committee
- Plaat, A.; Mostaghim, S.; Giannakoglou, K.; Stein, B. van; Bonsangue, M.M.
- Qualification
- Doctor (dr.)
- Awarding Institution
- Leiden Institute of Advanced Computer Science (LIACS) , Faculty of Science , Leiden University
- Date
- 2020-07-02