Title :
Bidding with memory in the presence of synergies: a genetic algorithm implementation
Author :
Mochon, A. ; Saez, Y. ; Quintana, D. ; Isasi, P.
Author_Institution :
UNED, Madrid
Abstract :
A genetic algorithm has been developed to solve bidding strategies in a dynamic multi-unit auction: the Ausubel auction. The genetic algorithm aims to maximize each bidder´s payoff. To this end, a memory system about past experiences has been implemented. An extensive set of experiments have been carried out where different parameters of the genetic algorithm have been used in order to make a robust test bed. The present model has been studied for several environments that involve the presence or absence of synergies. For each environment, the bidding strategies, along with their effects on revenue and efficiency, are analyzed. No theoretical predictions have been developed yet for this auction format when values involve synergies; therefore, the aim of this work is to study the auction outcome where theoretical predictions are unknown. The results obtained with the genetic algorithm developed in this research reveal that without synergies, bidders tend to bid sincerely. Nevertheless, in the presence of synergies, when bidders have memory about their past results, they tend to shade their bids.
Keywords :
commerce; genetic algorithms; Ausubel auction; bidder payoff maximization; bidding strategies; dynamic multiunit auction; genetic algorithm; memory system; revenue; Genetic algorithms;
Conference_Titel :
Evolutionary Computation, 2007. CEC 2007. IEEE Congress on
Conference_Location :
Singapore
Print_ISBN :
978-1-4244-1339-3
Electronic_ISBN :
978-1-4244-1340-9
DOI :
10.1109/CEC.2007.4424476