2026-07-19T22:34:54Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:4697 2024-03-06T10:35:15Z ddc:004 open_access

Reaching Consensus via Non-Bayesian Asynchronous Learning in Social Networks Feldman, Michal Immorlica, Nicole Lucier, Brendan Weinberg, S. Matthew Information Cascades Social Networks non-Bayesian Asynchronous Learning Expander Graphs Stochastic Processes We study the outcomes of information aggregation in online social networks. Our main result is that networks with certain realistic structural properties avoid information cascades and enable a population to effectively aggregate information. In our model, each individual in a network holds a private, independent opinion about a product or idea, biased toward a ground truth. Individuals declare their opinions asynchronously, can observe the stated opinions of their neighbors, and are free to update their declarations over time. Supposing that individuals conform with the majority report of their neighbors, we ask whether the population will eventually arrive at consensus on the ground truth. We show that the answer depends on the network structure: there exist networks for which consensus is unlikely, or for which declarations converge on the incorrect opinion with positive probability. On the other hand, we prove that for networks that are sparse and expansive, the population will converge to the correct opinion with high probability. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Michal Feldman and Nicole Immorlica and Brendan Lucier and S. Matthew Weinberg 2014 Is Part Of LIPIcs, Volume 28, Approximation, Randomization, and Combinatorial Optimization. Algorithms and Techniques (APPROX/RANDOM 2014) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.APPROX-RANDOM.2014.192 urn:nbn:de:0030-drops-46976 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.APPROX-RANDOM.2014.192 eng https://creativecommons.org/licenses/by/3.0/legalcode