View Item 
        •   Utrecht University Student Theses Repository Home
        • UU Theses Repository
        • Theses
        • View Item
        •   Utrecht University Student Theses Repository Home
        • UU Theses Repository
        • Theses
        • View Item
        JavaScript is disabled for your browser. Some features of this site may not work without it.

        Browse

        All of UU Student Theses RepositoryBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

        Evaluating the effectiveness and applicability of Novelty Search and other Quality-Diversity algorithms compared to niching

        Thumbnail
        View/Open
        Thesis.pdf (553.8Kb)
        Publication date
        2020
        Author
        Schonewille, M.
        Metadata
        Show full item record
        Summary
        Novelty Search is a promising new evolutionary algorithm, which claims to outperform traditional evolutionary algorithms in some cases. The idea is that sometimes, pursuing the objective, like in traditional evolutionary algorithms, may prevent the objective from being reached. In these cases, it might be better to explore solutions that are inherently different than previous ones instead of solutions that have a higher fitness value. Imagine walking through a maze towards a goal, but first having to walk away from it to eventually reach it. The question is how effective this method is compared to niching algorithms, another methods that tries to increase both quality and diversity. In this paper we will subject novelty search and niching algorithms to the classic maze experiment, in order to find out exactly how effective novelty search is when compared to niching and in what cases.
        URI
        https://studenttheses.uu.nl/handle/20.500.12932/37762
        Collections
        • Theses
        Utrecht university logo