Observational and reinforcement pattern learning: an exploratory study

Simple item page
dc.contributor.authorHanaki, N.
dc.contributor.authorKirman, A.
dc.contributor.authorPezanis-Christou, P.
dc.date.issued2018
dc.description.abstractUnderstanding how individuals learn in an unknown environment is an important problem in economics. We model and examine experimentally behavior in a very simple multi-armed bandit framework in which participants do not know the inter-temporal payoff structure. We propose a baseline reinforcement learning model that allows for pattern-recognition and change in the strategy space. We also analyse three augmented versions that accommodate observational learning from the actions and/or payoffs of another player. The models successfully reproduce the distributional properties of observed discovery times and total payoffs. Our study further shows that when one of the pair discovers the hidden pattern, observing another’s actions and/or payoffs improves discovery time compared to the baseline case.
dc.description.statementofresponsibilityNobuyuki Hanaki, Alan Kirman, Paul Pezanis-Christou
dc.identifier.citationEuropean Economic Review, 2018; 104:1-21
dc.identifier.doi10.1016/j.euroecorev.2018.01.009
dc.identifier.issn0014-2921
dc.identifier.issn1873-572X
dc.identifier.orcidPezanis-Christou, P. [0000-0001-6521-4139]
dc.identifier.urihttps://hdl.handle.net/2440/132302
dc.language.isoen
dc.publisherElsevier BV
dc.relation.granthttp://purl.org/au-research/grants/arc/DP140102949
dc.rights© 2018 Published by Elsevier B.V.
dc.source.urihttps://doi.org/10.1016/j.euroecorev.2018.01.009
dc.subjectMulti-armed bandit; reinforcement learning; payoff patterns; observational learning
dc.titleObservational and reinforcement pattern learning: an exploratory study
dc.typeJournal article
pubs.publication-statusPublished

Files

Collections

Economics publications