Bitte benutzen Sie diese Referenz, um auf diese Ressource zu verweisen: doi:10.22028/D291-27350
Titel: Incentives in dynamic markets
Verfasser: Kodric, Bojana
Sprache: Englisch
Erscheinungsjahr: 2017
DDC-Sachgruppe: 500 Naturwissenschaften
Dokumentart : Dissertation
Kurzfassung: In this thesis, we consider a variety of combinatorial optimization problems within a common theme of uncertainty and selfish behavior. In our first scenario, the input is collected from selfish players. Here, we study extensions of the so-called smoothness framework for mechanisms, a very useful technique for bounding the inefficiency of equilibria, to the cases of varying mechanism availability and participation of risk-averse players. In both of these cases, our main results are general theorems for the class of (lambda,mu)-smooth mechanisms. We show that these mechanisms guarantee at most a (small) constant factor performance loss in the extended settings. In our second scenario, we do not have access to the exact numerical input. Within this context, we explore combinatorial extensions of the well-known secretary problem under the assumption that the incoming elements only reveal their ordinal position within the set of previously arrived elements. We first observe that many existing algorithms for special matroid structures maintain their competitive ratio in the ordinal model. In contrast, we provide a lower bound for algorithms that are oblivious to the matroid structure. Finally, we design new algorithms that obtain constant competitive ratios for a variety of combinatorial problems.
Link zu diesem Datensatz: urn:nbn:de:bsz:291-scidok-ds-273509
Erstgutachter: Hoefer, Martin
Tag der mündlichen Prüfung: 23-Jul-2018
SciDok-Publikation: 24-Sep-2018
Fakultät: MI - Fakultät für Mathematik und Informatik
Fachrichtung: MI - Informatik
Fakultät / Institution:SciDok - Elektronische Dokumente der UdS

Dateien zu dieser Ressource:
Datei Beschreibung GrößeFormat 
thesis.pdf644,53 kBAdobe PDFÖffnen/Anzeigen

Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt.