Academic Journal
Online Regret Bounds for Satisficing in Markov Decision Processes.
| Τίτλος: | Online Regret Bounds for Satisficing in Markov Decision Processes. |
|---|---|
| Συγγραφείς: | Hajiabolhassan, Hossein1 (AUTHOR) hossein.hajiabolhassan@medunigraz.at, Ortner, Ronald2 (AUTHOR) rortner@unileoben.ac.at |
| Πηγή: | Mathematics of Operations Research (INFORMS). May2026, Vol. 51 Issue 2, p1097-1119. 23p. |
| Θεματικοί όροι: | *Markov processes, *Decision making, Reinforcement learning |
| Περίληψη: | We consider general reinforcement learning under the average reward criterion in Markov decision processes (MDPs), when the learner's goal is not to learn an optimal policy, but accepts any policy whose average reward is above a given satisfaction level σ. We show that with this more modest objective, it is possible to give algorithms that only have constant regret with respect to the level σ , provided that there is a policy above this level. This is a generalization of known results from the bandit setting to MDPs. Further, we present a more general algorithm that achieves the best of both worlds: If the optimal policy has average reward above σ , this algorithm has bounded regret with respect to σ. On the other hand, if all policies are below σ , then the expected regret with respect to the optimal policy is bounded as for the UCRL2 algorithm. Funding: Financial support from the Austrian Science Fund (FWF) [Grant TAI 590-N] is gratefully acknowledged. [ABSTRACT FROM AUTHOR] |
| Copyright of Mathematics of Operations Research (INFORMS) is the property of INFORMS: Institute for Operations Research & the Management Sciences and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Βάση Δεδομένων: | Business Source Index |
| FullText | Text: Availability: 0 CustomLinks: – Url: http://www.jstor.org/openurl?issn=0364765X&date=2026&volume=51&issue=2&spage=1097 Name: JSTOR (all content) (s7799221) Category: fullText Text: Icon: https://imageserver.ebscohost.com/branding/jstor/iconJSTOR.gif MouseOverText: Full Text from JSTOR |
|---|---|
| Header | DbId: bsx DbLabel: Business Source Index An: 193594469 RelevancyScore: 1492 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 1491.61413574219 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Online Regret Bounds for Satisficing in Markov Decision Processes. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Hajiabolhassan%2C+Hossein%22">Hajiabolhassan, Hossein</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> hossein.hajiabolhassan@medunigraz.at</i><br /><searchLink fieldCode="AR" term="%22Ortner%2C+Ronald%22">Ortner, Ronald</searchLink><relatesTo>2</relatesTo> (AUTHOR)<i> rortner@unileoben.ac.at</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Mathematics+of+Operations+Research+%28INFORMS%29%22">Mathematics of Operations Research (INFORMS)</searchLink>. May2026, Vol. 51 Issue 2, p1097-1119. 23p. – Name: Subject Label: Subject Terms Group: Su Data: *<searchLink fieldCode="DE" term="%22Markov+processes%22">Markov processes</searchLink><br />*<searchLink fieldCode="DE" term="%22Decision+making%22">Decision making</searchLink><br /><searchLink fieldCode="DE" term="%22Reinforcement+learning%22">Reinforcement learning</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: We consider general reinforcement learning under the average reward criterion in Markov decision processes (MDPs), when the learner's goal is not to learn an optimal policy, but accepts any policy whose average reward is above a given satisfaction level σ. We show that with this more modest objective, it is possible to give algorithms that only have constant regret with respect to the level σ , provided that there is a policy above this level. This is a generalization of known results from the bandit setting to MDPs. Further, we present a more general algorithm that achieves the best of both worlds: If the optimal policy has average reward above σ , this algorithm has bounded regret with respect to σ. On the other hand, if all policies are below σ , then the expected regret with respect to the optimal policy is bounded as for the UCRL2 algorithm. Funding: Financial support from the Austrian Science Fund (FWF) [Grant TAI 590-N] is gratefully acknowledged. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Mathematics of Operations Research (INFORMS) is the property of INFORMS: Institute for Operations Research & the Management Sciences and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=bsx&AN=193594469 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1287/moor.2023.0275 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 23 StartPage: 1097 Subjects: – SubjectFull: Markov processes Type: general – SubjectFull: Decision making Type: general – SubjectFull: Reinforcement learning Type: general Titles: – TitleFull: Online Regret Bounds for Satisficing in Markov Decision Processes. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Hajiabolhassan, Hossein – PersonEntity: Name: NameFull: Ortner, Ronald IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 05 Text: May2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 0364765X Numbering: – Type: volume Value: 51 – Type: issue Value: 2 Titles: – TitleFull: Mathematics of Operations Research (INFORMS) Type: main |
| ResultId | 1 |