StreamingBandit : Experimenting with Bandit Policies

Show simple item record

dc.contributor.author Kruijswijk, Jules
dc.contributor.author van Emden, Robin
dc.contributor.author Parvinen, Petri
dc.contributor.author Kaptein, Maurits
dc.date.accessioned 2021-01-13T09:16:01Z
dc.date.available 2021-01-13T09:16:01Z
dc.date.issued 2020-08
dc.identifier.citation Kruijswijk , J , van Emden , R , Parvinen , P & Kaptein , M 2020 , ' StreamingBandit : Experimenting with Bandit Policies ' , Journal of Statistical Software , vol. 94 , no. 9 , pp. 1-47 . https://doi.org/10.18637/jss.v094.i09
dc.identifier.other PURE: 158950485
dc.identifier.other PURE UUID: 7dcc717c-ec09-4046-aa08-c97fc6d7c608
dc.identifier.other WOS: 000565254700001
dc.identifier.other ORCID: /0000-0002-9977-4975/work/86940281
dc.identifier.uri http://hdl.handle.net/10138/324462
dc.description.abstract A large number of statistical decision problems in the social sciences and beyond can be framed as a (contextual) multi-armed bandit problem. However, it is notoriously hard to develop and evaluate policies that tackle these types of problems, and to use such policies in applied studies. To address this issue, this paper introduces StreamingBandit, a Python web application for developing and testing bandit policies in field studies. StreamingBandit can sequentially select treatments using (online) policies in real time. Once StreamingBandit is implemented in an applied context, different policies can be tested, altered, nested, and compared. StreamingBandit makes it easy to apply a multitude of bandit policies for sequential allocation in field experiments, and allows for the quick development and re-use of novel policies. In this article, we detail the implementation logic of StreamingBandit and provide several examples of its use. en
dc.format.extent 47
dc.language.iso eng
dc.relation.ispartof Journal of Statistical Software
dc.rights cc_by
dc.rights.uri info:eu-repo/semantics/openAccess
dc.subject sequential decision-making
dc.subject multi-armed bandit
dc.subject data streams
dc.subject sequential experimentation
dc.subject Python
dc.subject CLINICAL-TRIALS
dc.subject 112 Statistics and probability
dc.title StreamingBandit : Experimenting with Bandit Policies en
dc.type Article
dc.contributor.organization Department of Economics and Management
dc.contributor.organization Department of Forest Sciences
dc.contributor.organization Forest Economics, Business and Society
dc.description.reviewstatus Peer reviewed
dc.relation.doi https://doi.org/10.18637/jss.v094.i09
dc.relation.issn 1548-7660
dc.rights.accesslevel openAccess
dc.type.version publishedVersion

Files in this item

Total number of downloads: Loading...

Files Size Format View
v94i09.pdf 1013.Kb PDF View/Open

This item appears in the following Collection(s)

Show simple item record