Guiding evolutionary search towards innovative solutions

Leiden Repository

Guiding evolutionary search towards innovative solutions

Title: Guiding evolutionary search towards innovative solutions
Author: Reehuis, Edgar Julius Jeronimus
Publisher: Leiden Institute of Advanced Computer Science (LIACS), Natural Computing Group, Faculty of Science, Leiden University
Issue Date: 2013-12-17
Keywords: Design optimization
Surrogate models
Novelty
Interestingness
Aerodynamic simulation
Evolutionary algorithms
Abstract: The main goal of this work is to develop a method that, operating on top of an Evolutionary Algorithm, increases its likeliness of finding innovative solutions. This likeliness is laid out to be increased with the diversity of the solutions found, provided that they are of sufficient quality. The developed method needs to be applicable in a scenario in which the search is required to be started from a single, fixed solution. Therefore, a scheme is envisioned in which the search is performed in a sequential fashion, zooming in on a locally-optimal solution, and then exploring for a new potentially high-quality region based on a memory of solutions encountered earlier in the search. Two exploration criteria, one using an archive of earlier solutions as memory and the other deriving from a surrogate model trained on earlier solutions, were established to be worthwhile for integration into quality-based search. The resulting schemes were applied to a real-world airfoil optimization task, showing both to perform better than the baseline method of multiple standard optimization runs. The model-based approach delivers the best results, in the sense that it finds more solutions, more diverse solutions, and better-quality solutions than the baseline method.
Description: Promotores: T.H.W. Bäck, B.A. Sendhoff
With summary in Dutch
Faculty: Faculteit der Wiskunde en Natuurwetenschappen
Citation: Reehuis, E.J.J., 2013, Doctoral thesis, Leiden University
Sponsor: Honda Research Institute Europe (HRI-EU)
Handle: http://hdl.handle.net/1887/22875
 

Files in this item

Description Size View
application/pdf Full text 27.04Mb View/Open
application/pdf Cover 13.85Mb View/Open
application/pdf Title page_Contents 166.2Kb View/Open
application/pdf Chapter 1 Introduction 199.2Kb View/Open
application/pdf Chapter 2 545.3Kb View/Open
application/pdf Chapter 3 575.8Kb View/Open
application/pdf Chapter 4 13.49Mb View/Open
application/pdf Chapter 5 5.992Mb View/Open
application/pdf Chapter 6 6.722Mb View/Open
application/pdf Chapter 7 Conclusions and outlook 203.2Kb View/Open
application/pdf Bibliography 188.3Kb View/Open
application/pdf Summary in Dutch 155.9Kb View/Open
application/pdf Acknowledgements_About the author 124.4Kb View/Open
application/pdf Propositions 106.5Kb View/Open

This item appears in the following Collection(s)