Guido Schaefer

Networks and Optimization group
Centrum Wiskunde & Informatica
Department of Econometrics and Operations Research
Vrije Universiteit Amsterdam.

PRESENTATION

Bio:
My background is Operations Research, in particular Combinatorial Optimization and Algorithmic Game Theory. 
I am very much intrigued by the often counterintuitive phenomena (such as the Braess Paradox or the Tragedy of the Commons) that arise as a result of strategic decision making in multi-agent systems or networks. In my research, I focus on the modeling and analysis of such phenomena with the ultimate goal to design incentives to avoid or at least reduce the negative effects of selfish behavior. 


Topic:
Game theory provides a powerful toolbox to study situations of strategic decision making on mathematical grounds. Sometimes, however, the conclusions at which we arrive seem to be off or at least counterintuitive when compared to our experiences from reality (or observations made in experiments). As an example, our recent studies of a fundamental class of congestion games seem to suggest that altruistic behavior might be harmful for society (in a certain mathematical sense), while spiteful behavior might be beneficial. 

It is easy to brush such findings away simply by questioning the appropriateness of the (simplistic) model or solution concepts being used. And, certainly, this should motivate us to reconsider our model and refine our analysis to arrive at a precise understanding. But what if there is more to this? I would be curious to discuss these phenomena and their consequences and learn whether such counterintuitive effects have been encountered in the social (network) sciences. Also I would be interested to hear about new network applications that are amenable to game-theoretical analyses.