Learning Agents for Iterative Voting

Airiau, Stéphane; Grandi, Umberto; Studzinski Perotto, Filipo

Airiau, Stéphane; Grandi, Umberto; Studzinski Perotto, Filipo (2017), Learning Agents for Iterative Voting, dans Rothe, Jörg, Algorithmic Decision Theory - 5th International Conference (ADT 2017), Springer International Publishing : Berlin Heidelberg, p. 139-152. 10.1007/978-3-319-67504-6_10

Type

Communication / Conférence

Date

2017

Titre du colloque

5th International Conference (ADT 2017)

Date du colloque

2017-10

Ville du colloque

Luxembourg

Pays du colloque

Luxembourg

Titre de l'ouvrage

Algorithmic Decision Theory - 5th International Conference (ADT 2017)

Auteurs de l’ouvrage

Rothe, Jörg

Éditeur

Springer International Publishing

Ville d’édition

Berlin Heidelberg

Isbn

978-3-319-67503-9

Nombre de pages

390

Pages

139-152

Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Grandi, Umberto

Studzinski Perotto, Filipo

Résumé (EN)

This paper assesses the learning capabilities of agents in a situation of collective choice. Each agent is endowed with a private preference concerning a number of alternative candidates, and participates in an iterated plurality election. Agents get rewards depending on the winner of each election, and adjust their voting strategy using reinforcement learning. By conducting extensive simulations, we show that our agents are capable of learning how to take decisions at the level of well-known voting procedures, and that these decisions maintain good choice-theoretic properties when increasing the number of agents or candidates.

Mots-clés

Computational social choice; Iterative voting; Bandit algorithms