StreamingBandit : Experimenting with Bandit Policies

Visa fullständig post



Permalänk

http://hdl.handle.net/10138/324462

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

Titel: StreamingBandit : Experimenting with Bandit Policies
Författare: Kruijswijk, Jules; van Emden, Robin; Parvinen, Petri; Kaptein, Maurits
Upphovmannens organisation: Department of Economics and Management
Department of Forest Sciences
Forest Economics, Business and Society
Datum: 2020-08
Språk: eng
Sidantal: 47
Tillhör serie: Journal of Statistical Software
ISSN: 1548-7660
DOI: https://doi.org/10.18637/jss.v094.i09
Permanenta länken (URI): http://hdl.handle.net/10138/324462
Abstrakt: 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.
Subject: sequential decision-making
multi-armed bandit
data streams
sequential experimentation
Python
CLINICAL-TRIALS
112 Statistics and probability
Referentgranskad: Ja
Licens: cc_by
Användningsbegränsning: openAccess
Parallelpublicerad version: publishedVersion


Filer under denna titel

Totalt antal nerladdningar: Laddar...

Filer Storlek Format Granska
v94i09.pdf 1013.Kb PDF Granska/Öppna

Detta dokument registreras i samling:

Visa fullständig post