Learning to Infer Structures of Network Games

Emanuele Rossi, Federico Monti, Yan Leng, Michael Bronstein, Xiaowen Dong
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:18809-18827, 2022.

Abstract

Strategic interactions between a group of individuals or organisations can be modelled as games played on networks, where a player’s payoff depends not only on their actions but also on those of their neighbours. Inferring the network structure from observed game outcomes (equilibrium actions) is an important problem with numerous potential applications in economics and social sciences. Existing methods mostly require the knowledge of the utility function associated with the game, which is often unrealistic to obtain in real-world scenarios. We adopt a transformer-like architecture which correctly accounts for the symmetries of the problem and learns a mapping from the equilibrium actions to the network structure of the game without explicit knowledge of the utility function. We test our method on three different types of network games using both synthetic and real-world data, and demonstrate its effectiveness in network structure inference and superior performance over existing methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-rossi22a, title = {Learning to Infer Structures of Network Games}, author = {Rossi, Emanuele and Monti, Federico and Leng, Yan and Bronstein, Michael and Dong, Xiaowen}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {18809--18827}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/rossi22a/rossi22a.pdf}, url = {https://proceedings.mlr.press/v162/rossi22a.html}, abstract = {Strategic interactions between a group of individuals or organisations can be modelled as games played on networks, where a player’s payoff depends not only on their actions but also on those of their neighbours. Inferring the network structure from observed game outcomes (equilibrium actions) is an important problem with numerous potential applications in economics and social sciences. Existing methods mostly require the knowledge of the utility function associated with the game, which is often unrealistic to obtain in real-world scenarios. We adopt a transformer-like architecture which correctly accounts for the symmetries of the problem and learns a mapping from the equilibrium actions to the network structure of the game without explicit knowledge of the utility function. We test our method on three different types of network games using both synthetic and real-world data, and demonstrate its effectiveness in network structure inference and superior performance over existing methods.} }
Endnote
%0 Conference Paper %T Learning to Infer Structures of Network Games %A Emanuele Rossi %A Federico Monti %A Yan Leng %A Michael Bronstein %A Xiaowen Dong %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-rossi22a %I PMLR %P 18809--18827 %U https://proceedings.mlr.press/v162/rossi22a.html %V 162 %X Strategic interactions between a group of individuals or organisations can be modelled as games played on networks, where a player’s payoff depends not only on their actions but also on those of their neighbours. Inferring the network structure from observed game outcomes (equilibrium actions) is an important problem with numerous potential applications in economics and social sciences. Existing methods mostly require the knowledge of the utility function associated with the game, which is often unrealistic to obtain in real-world scenarios. We adopt a transformer-like architecture which correctly accounts for the symmetries of the problem and learns a mapping from the equilibrium actions to the network structure of the game without explicit knowledge of the utility function. We test our method on three different types of network games using both synthetic and real-world data, and demonstrate its effectiveness in network structure inference and superior performance over existing methods.
APA
Rossi, E., Monti, F., Leng, Y., Bronstein, M. & Dong, X.. (2022). Learning to Infer Structures of Network Games. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:18809-18827 Available from https://proceedings.mlr.press/v162/rossi22a.html.

Related Material