A General Framework for Rational Learning in Social Networks

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. This paper provides a formal characterization of the process of rational learning in social networks. Agents receive initial private information and select an action out of a choice set under uncertainty in each of infinitely many periods, observing the history of choices of their neighbors. Choices are made based on a common behavioral rule. Conditions under which rational learning leads to global consensus, local indifference, and local disagreement are characterized. In the general setting considered, rational learning can lead to pairs of neighbors selecting different actions once learning ends while not being indifferent among their choices. Terms of use: Documents in EconStor mayThe effect of the network structure on the degree of information aggregation and speed of convergence is also considered, and an answer to the question of optimal information aggregation in networks is provided. The results highlight distinguishing features between properties of Bayesian and non-Bayesian learning in social networks.

Bayesian Decision Making in Groups is Hard

“…Our results are related to the line of work that studies conditions for consensus and learning among rational agents (Mossel et al 2018, Mueller-Frank 2013, Smith and Sørensen 2000. Consensus refers to all agents converging in their actions or belief (cf.…”

TICE, mémoire et compréhension de texte scientifique en français L2

“…Avancées bien utiles à l'heure de la mondialisation et de la généralisation de l'utilisation d'internet et des réseaux sociaux (Mueller-Frank, 2013). Les recherches sur les littératies numériques et plurilingues modifient en effet les rapports entre cognition, mémoire, multimédia et apprentissage et imposent de prendre en compte les contextes plurilingues et pluriculturels des apprenants (Bounouara, 2011 ;Bounouara & Legros, 2012, 2014Legros & Bounouara, 2010 ;Lyman-Hager, Johns, Nocon, & Davis, 2002).…”

Learning in network games with incomplete information: asymptotic analysis and tractable implementation of rational behavior

“…There exists an extensive literature on Bayesian learning over networks for scenarios without payoff externalities [13]- [16]. One may think of this problem as a variant of distributed estimation since agents intend to compute an estimate based on global information by aggregating local information and successively refining their estimates using those of their neighbors.…”

“…The interested readers can refer to "Example of Learning with Payoff Externality" for an illustration of the rational learning process with both payoff and informational externalities. An example of learning with only informational externalities is given in [16]. In this section, we introduce games with utility functions that are quadratic both in the state of the world and agents' actions.…”

“…By the symmetry property of the utility function strategic complementarity implies that unilateral deviations by i and j to each other's actions are at least as good as playing their own limit actions in expectation. However, by the imitation principle in (16), this behavior can never yield strictly higher payoffs. Hence, it must be the case that deviations to neighbors' limit actions result in no change in the expected performance of agents.…”