Cui, Kai ; Dayanıklı, Gökçe ; Laurière, Mathieu ; Geist, Matthieu ; Pietquin, Olivier ; Koeppl, Heinz (2024)
Learning Discrete-Time Major-Minor Mean Field Games.
38th AAAI Conference on Artificial Intelligence. Vancouver, Canada (20.02.2024 - 27.02.2024)
doi: 10.26083/tuprints-00028687
Konferenzveröffentlichung, Zweitveröffentlichung, Postprint
 | Es ist eine neuere Version dieses Eintrags verfügbar. |
Kurzbeschreibung (Abstract)
Recent techniques based on Mean Field Games (MFGs) allow the scalable analysis of multi-player games with many similar, rational agents. However, standard MFGs remain limited to homogeneous players that weakly influence each other, and cannot model major players that strongly influence other players, severely limiting the class of problems that can be handled. We propose a novel discrete time version of major-minor MFGs (M3FGs), along with a learning algorithm based on fictitious play and partitioning the probability simplex. Importantly, M3FGs generalize MFGs with common noise and can handle not only random exogeneous environment states but also major players. A key challenge is that the mean field is stochastic and not deterministic as in standard MFGs. Our theoretical investigation verifies both the M3FG model and its algorithmic solution, showing firstly the well-posedness of the M3FG model starting from a finite game of interest, and secondly convergence and approximation guarantees of the fictitious play algorithm. Then, we empirically verify the obtained theoretical results, ablating some of the theoretical assumptions made, and show successful equilibrium learning in three example problems. Overall, we establish a learning framework for a novel and broad class of tractable games.
| Typ des Eintrags: | Konferenzveröffentlichung |
|---|---|
| Erschienen: | 2024 |
| Autor(en): | Cui, Kai ; Dayanıklı, Gökçe ; Laurière, Mathieu ; Geist, Matthieu ; Pietquin, Olivier ; Koeppl, Heinz |
| Art des Eintrags: | Zweitveröffentlichung |
| Titel: | Learning Discrete-Time Major-Minor Mean Field Games |
| Sprache: | Englisch |
| Publikationsjahr: | 16 Dezember 2024 |
| Ort: | Darmstadt |
| Publikationsdatum der Erstveröffentlichung: | 2024 |
| Ort der Erstveröffentlichung: | Menlo Park, Calif. |
| Verlag: | AAAI |
| Titel der Zeitschrift, Zeitung oder Schriftenreihe: | Proceedings of the AAAI Conference on Artificial Intelligence |
| Jahrgang/Volume einer Zeitschrift: | 38 |
| (Heft-)Nummer: | 9 |
| Buchtitel: | Proceedings of the 38th AAAI Conference on Artificial Intelligence |
| Veranstaltungstitel: | 38th AAAI Conference on Artificial Intelligence |
| Veranstaltungsort: | Vancouver, Canada |
| Veranstaltungsdatum: | 20.02.2024 - 27.02.2024 |
| DOI: | 10.26083/tuprints-00028687 |
| URL / URN: | https://tuprints.ulb.tu-darmstadt.de/28687 |
| Zugehörige Links: | |
| Herkunft: | Zweitveröffentlichungsservice |
| Kurzbeschreibung (Abstract): | Recent techniques based on Mean Field Games (MFGs) allow the scalable analysis of multi-player games with many similar, rational agents. However, standard MFGs remain limited to homogeneous players that weakly influence each other, and cannot model major players that strongly influence other players, severely limiting the class of problems that can be handled. We propose a novel discrete time version of major-minor MFGs (M3FGs), along with a learning algorithm based on fictitious play and partitioning the probability simplex. Importantly, M3FGs generalize MFGs with common noise and can handle not only random exogeneous environment states but also major players. A key challenge is that the mean field is stochastic and not deterministic as in standard MFGs. Our theoretical investigation verifies both the M3FG model and its algorithmic solution, showing firstly the well-posedness of the M3FG model starting from a finite game of interest, and secondly convergence and approximation guarantees of the fictitious play algorithm. Then, we empirically verify the obtained theoretical results, ablating some of the theoretical assumptions made, and show successful equilibrium learning in three example problems. Overall, we establish a learning framework for a novel and broad class of tractable games. |
| Status: | Postprint |
| URN: | urn:nbn:de:tuda-tuprints-286878 |
| Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 000 Allgemeines, Informatik, Informationswissenschaft > 004 Informatik 600 Technik, Medizin, angewandte Wissenschaften > 621.3 Elektrotechnik, Elektronik |
| Fachbereich(e)/-gebiet(e): | 18 Fachbereich Elektrotechnik und Informationstechnik 18 Fachbereich Elektrotechnik und Informationstechnik > Self-Organizing Systems Lab |
| Hinterlegungsdatum: | 16 Dez 2024 14:02 |
| Letzte Änderung: | 16 Jan 2025 15:19 |
| PPN: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
Verfügbare Versionen dieses Eintrags
- Learning Discrete-Time Major-Minor Mean Field Games. (deposited 16 Dez 2024 14:02) [Gegenwärtig angezeigt]
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum