Q-learning in regularized mean-field games

buir.contributor.authorSaldi, Naci
buir.contributor.orcidSaldi, Naci|0000-0002-2677-7366
dc.citation.epage29en_US
dc.citation.spage1en_US
dc.contributor.authorAnahtarci, B.
dc.contributor.authorKariksiz, C.D.
dc.contributor.authorSaldi, Naci
dc.date.accessioned2023-02-16T10:30:13Z
dc.date.available2023-02-16T10:30:13Z
dc.date.issued2022-05-23
dc.departmentDepartment of Mathematicsen_US
dc.description.abstractIn this paper, we introduce a regularized mean-field game and study learning of this game under an infinite-horizon discounted reward function. Regularization is introduced by adding a strongly concave regularization function to the one-stage reward function in the classical mean-field game model. We establish a value iteration based learning algorithm to this regularized mean-field game using fitted Q-learning. The regularization term in general makes reinforcement learning algorithm more robust to the system components. Moreover, it enables us to establish error analysis of the learning algorithm without imposing restrictive convexity assumptions on the system components, which are needed in the absence of a regularization term.en_US
dc.identifier.doi10.1007/s13235-022-00450-2en_US
dc.identifier.eissn2153-0793
dc.identifier.issn2153-0785
dc.identifier.urihttp://hdl.handle.net/11693/111428
dc.language.isoEnglishen_US
dc.publisherBirkhaeuser Scienceen_US
dc.relation.isversionofhttps://www.doi.org/10.1007/s13235-022-00450-2en_US
dc.source.titleDynamic Games and Applicationsen_US
dc.subjectMean-field gamesen_US
dc.subjectQ-learningen_US
dc.subjectRegularized Markov decision processesen_US
dc.subjectDiscounted rewarden_US
dc.titleQ-learning in regularized mean-field gamesen_US
dc.typeArticleen_US
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Q-learning in regularized mean-field games.pdf
Size:
689.12 KB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.69 KB
Format:
Item-specific license agreed upon to submission
Description: