Computational social choice
Our scientific interests regard computational social choice, a new area that connects computer science and economics. We design algorithms for, study computational complexity of, and analyze axiomatic properties of various election rules, focusing on committee elections. Perhaps the best known example of a committee election is choosing a country’s parliament. Here the goal is to choose a committee that in some sense would proportionally represent the views of the voters. Other examples include choosing finalists of competitions (where the goal is to select individually excellent candidates) or choosing items that an Internet store puts on its homepage (where the goal is to choose as diverse a „committee” of items as possible, so that as many customers as possible find the store’s offering attractive).
Within our research, we see computational complexity results for the problem of computing winning committees under various rules and, as these problems typical are NP-hard, we see approximation algorithms, FPT results, and heuristic methods for them. We are also interested in axiomatic properties of multiwinner voting rules, in the problem of analyzing candidates’ performance in elections, in the problems of manipulating election results, and in comparing elections.
Keywords: algorithmics, theory of computational complexity
Contact: Piotr Faliszewski
Research team: CSG
Team leader: Prof. Jacek Kitowski