Linear Partial Monitoring for Sequential Decision Making: Algorithms, Regret Bounds and Applications
dc.contributor.author
Kirschner, Johannes
dc.contributor.author
Lattimore, Tor
dc.contributor.author
Krause, Andreas
dc.date.accessioned
2024-02-26T11:00:44Z
dc.date.available
2024-02-25T08:17:31Z
dc.date.available
2024-02-26T11:00:44Z
dc.date.issued
2023
dc.identifier.issn
1532-4435
dc.identifier.issn
1533-7928
dc.identifier.uri
http://hdl.handle.net/20.500.11850/661344
dc.identifier.doi
10.3929/ethz-b-000661344
dc.description.abstract
Partial monitoring is an expressive framework for sequential decision-making with an abundance of applications, including graph-structured and dueling bandits, dynamic pricing and transductive feedback models. We survey and extend recent results on the linear formulation of partial monitoring that naturally generalizes the standard linear bandit setting. The main result is that a single algorithm, information-directed sampling (IDS), is (nearly) worst-case rate optimal in all finite-action games. We present a simple and unified analysis of stochastic partial monitoring, and further extend the model to the contextual and kernelized setting.
en_US
dc.format
application/pdf
en_US
dc.language.iso
en
en_US
dc.publisher
Microtome Publishing
en_US
dc.rights.uri
http://creativecommons.org/licenses/by/4.0/
dc.subject
Sequential Decision-Making
en_US
dc.subject
Linear Partial Monitoring
en_US
dc.subject
Information-Directed Sampling
en_US
dc.subject
Linear Bandits
en_US
dc.title
Linear Partial Monitoring for Sequential Decision Making: Algorithms, Regret Bounds and Applications
en_US
dc.type
Journal Article
dc.rights.license
Creative Commons Attribution 4.0 International
dc.date.published
2023-08
ethz.journal.title
Journal of Machine Learning Research
ethz.journal.volume
24
en_US
ethz.journal.abbreviated
J. Mach. Learn. Res.
ethz.pages.start
346
en_US
ethz.size
45 p.
en_US
ethz.version.deposit
publishedVersion
en_US
ethz.grant
Reliable Data-Driven Decision Making in Cyber-Physical Systems
en_US
ethz.identifier.wos
ethz.publication.status
published
en_US
ethz.identifier.url
https://www.jmlr.org/papers/v24/22-1248.html
ethz.grant.agreementno
815943
ethz.grant.fundername
EC
ethz.grant.funderDoi
10.13039/501100000780
ethz.grant.program
H2020
ethz.date.deposited
2024-02-25T08:17:35Z
ethz.source
WOS
ethz.eth
yes
en_US
ethz.availability
Open access
en_US
ethz.rosetta.installDate
2024-02-26T11:00:45Z
ethz.rosetta.lastUpdated
2024-02-26T11:00:45Z
ethz.rosetta.exportRequired
true
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Linear%20Partial%20Monitoring%20for%20Sequential%20Decision%20Making:%20Algorithms,%20Regret%20Bounds%20and%20Applications&rft.jtitle=Journal%20of%20Machine%20Learning%20Research&rft.date=2023&rft.volume=24&rft.spage=346&rft.issn=1532-4435&1533-7928&rft.au=Kirschner,%20Johannes&Lattimore,%20Tor&Krause,%20Andreas&rft.genre=article&
Files in this item
Publication type
-
Journal Article [130261]