Vés al contingut (premeu Retorn)

Conferència: Power indices and simple games -- an underrated technique in AI

Quan
02/03/2023 de 12:30 a 13:30
On
Sala de Conferències edifici TR5
Afegeix un esdeveniment al calendari
iCal

El Prof. Jochen Staudacher de la Kempten University of Applied Sciences visitarà a l'ESEIAAT.

La seva expertesa recau en la teoria de jocs i s'ha centrat en els jocs simples i els jocs de majoria ponderada. També a fet un software molt extens que calcula paràmetres, propietats i "conceptes solució" relacionats.

El dijous 2 de març de 12:30 a 13:30 farà una conferència a la Sala de Conferències del TR5 (Porta 1.41):

Speaker:  Jochen Staudacher (Faculty of Computer Science, Kempten University of Applied Sciences, Germany)

Title: Power indices and simple games -- an underrated technique in AI

Abstract of talk:
The talk gives an overview about power indices for simple games, i.e. games in which we have only winning coalitions with value 1 and losing coalitions with value 0. We explain why we regard these approaches underrated techniques in AI. We discuss efficient algorithms for power index computation as well as practical applications for financial, social or biological networks. Specifically, we look into the computation of power indices for weighted voting games using the paradigm of dynamic programming.

References:
[1] Jochen Staudacher, László  Á. Kóczy, Izabella Stach, Jan Filipp, Marcus Kramer, Till Noffke, Linus Olsson, Jonas Pichler, Tobias Singer (2021): Computing power indices for weighted voting games via dynamic programming, Operations Research and Decisions 31(2), pp. 123–145. DOI 10.37190/ord210206.
[2] Jochen Staudacher, Felix Wagner, Jan Filipp (2022): Dynamic Programming for Computing Power Indices for Weighted Voting Games with Precoalitions. Games 2022, 13(1), 6. 17 pages. DOI 10.3390/g13010006.
[3] Jochen Staudacher, Linus Olsson, Izabella Stach (2021): Implicit power indices for measuring indirect control in corporate structures. In: Nguyen N.T., Kowalczyk R., Mercik J., Motylska-Kuzma A. (eds.) Transactions on Computational Collective Intelligence XXXV. Lecture Notes in Computer Science, vol. No. 13010, pp. 73-93, Springer, Berlin, Heidelberg. DOI 10.1007/978-3-662-64563-5_4.