19 January 2023, Computational Social Choice Seminar, Christian Laußmann

Abstract

Participatory budgeting (PB) is a democratic tool for deciding how to spend an available amount of money: Do the citizens prefer spending the budget on the playground, and the new bike track, or are they more happy with spending the budget on refurbishing the museum? PB is used already for tasks like this. However, it is not suitable in many cases. While for small projects like a playground the cost and building time might be easy to estimate, the larger the projects are, the more uncertainty comes with their implementation. For instance, refurbishing a museum could turn out to cost way more than estimated. We look at a few examples from the real world to illustrate how wrong estimates can be. We then identify what kinds of uncertainty come with PB campaigns. Finally, we provide some first results on how we can deal with uncertainty algorithmically.

For more information on the Computational Social Choice Seminar, please consult https://staff.science.uva.nl/u.endriss/seminar/.