From Individual Goals to Collective Decisions
Arianna Novaro (IRIT - Université de Toulouse)13 déc. 2017 - 10:00
Classical rules for collective decision-making often require agents to fully specify their preference or opinion to compute the result. In particular, judgment aggregation rules require each agent to answer a yes/no question on a set of issues and they then output a collective judgment. In this talk I will relax this assumption by letting agents express their goals by means of propositional formulas on a finite set of binary issues. I will present some rules for aggregating individual goals into a decision for the group, as well as some adaptations to this setting of axiomatic properties from the literature on Social Choice Theory. Finally, I will discuss the problem of determining the outcome for the presented rules (i.e., the winner determination problem) from a computational complexity point of view.