Global bandits
| buir.contributor.author | Tekin, Cem | |
| dc.citation.epage | 5811 | en_US |
| dc.citation.issueNumber | 12 | en_US |
| dc.citation.spage | 5798 | en_US |
| dc.citation.volumeNumber | 29 | en_US |
| dc.contributor.author | Atan, O. | en_US |
| dc.contributor.author | Tekin, Cem | en_US |
| dc.contributor.author | Schaar, M. V. D. | en_US |
| dc.date.accessioned | 2019-02-21T16:05:50Z | |
| dc.date.available | 2019-02-21T16:05:50Z | |
| dc.date.issued | 2018 | en_US |
| dc.department | Department of Electrical and Electronics Engineering | en_US |
| dc.description.abstract | Multiarmed bandits (MABs) model sequential decision-making problems, in which a learner sequentially chooses arms with unknown reward distributions in order to maximize its cumulative reward. Most of the prior works on MAB assume that the reward distributions of each arm are independent. But in a wide variety of decision problems - from drug dosage to dynamic pricing - the expected rewards of different arms are correlated, so that selecting one arm provides information about the expected rewards of other arms as well. We propose and analyze a class of models of such decision problems, which we call global bandits (GB). In the case in which rewards of all arms are deterministic functions of a single unknown parameter, we construct a greedy policy that achieves bounded regret, with a bound that depends on the single true parameter of the problem. Hence, this policy selects suboptimal arms only finitely many times with probability one. For this case, we also obtain a bound on regret that is independent of the true parameter; this bound is sublinear, with an exponent that depends on the informativeness of the arms. We also propose a variant of the greedy policy that achieves O(√T) worst case and O(1) parameter-dependent regret. Finally, we perform experiments on dynamic pricing and show that the proposed algorithms achieve significant gains with respect to the well-known benchmarks. | |
| dc.description.provenance | Made available in DSpace on 2019-02-21T16:05:50Z (GMT). No. of bitstreams: 1 Bilkent-research-paper.pdf: 222869 bytes, checksum: 842af2b9bd649e7f548593affdbafbb3 (MD5) Previous issue date: 2018 | en |
| dc.description.sponsorship | Manuscript received April 13, 2017; revised December 21, 2017; accepted March 1, 2018. Date of publication April 12, 2018; date of current version November 16, 2018. The work of O. Atan and M. van der Schaar was supported by the NSF under Grant 1533983, Grant 1407712, and Grant 1462245. This paper was presented at the 2015 International Conference on Artificial Intelligence and Statistics (AISTATS), San Diego, CA, USA, May 2015. (Corresponding author: Onur Atan.) O. Atan and M. van der Schaar are with the Department of Electrical Engineering, University of California at Los Angeles, Los Angeles, CA 90095 USA (e-mail: oatan@ucla.edu; mihaela@ee.ucla.edu). | |
| dc.identifier.doi | 10.1109/TNNLS.2018.2818742 | |
| dc.identifier.issn | 2162-237X | |
| dc.identifier.uri | http://hdl.handle.net/11693/50276 | |
| dc.language.iso | English | |
| dc.publisher | Institute of Electrical and Electronics Engineers | |
| dc.relation.isversionof | https://doi.org/10.1109/TNNLS.2018.2818742 | |
| dc.relation.project | University of California, UC - National Science Foundation, NSF: 1462245 - National Science Foundation, NSF: 1533983 - National Science Foundation, NSF: 1407712 | |
| dc.source.title | IEEE Transactions on Neural Networks and Learning Systems | en_US |
| dc.subject | Bounded regret | en_US |
| dc.subject | Informative arms | en_US |
| dc.subject | Multiarmed bandits (MABs) | en_US |
| dc.subject | Online learning | en_US |
| dc.subject | Regret analysis | en_US |
| dc.title | Global bandits | en_US |
| dc.type | Article | en_US |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- Global_bandits.pdf
- Size:
- 1.76 MB
- Format:
- Adobe Portable Document Format
- Description:
- Full printable version