A0963
Title: Frequentist guarantees of distributed (non)-Bayesian inference
Authors: Cesar Augusto Uribe Meneses - Rice University (United States) [presenting]
Abstract: Motivated by the need to analyze large, decentralized datasets, distributed Bayesian inference has become a critical research area across multiple fields, including statistics, electrical engineering, and economics. Frequentist properties are established, such as posterior consistency, asymptotic normality, and posterior contraction rates, for the distributed (non-) Bayesian inference problem among agents connected via a communication network. Results show that, under appropriate assumptions on the communication graph, distributed Bayesian inference retains parametric efficiency while enhancing robustness in uncertainty quantification. The trade-off between statistical efficiency and communication efficiency is also explored by examining how the design and size of the communication graph impact the posterior contraction rate. Furthermore, the analysis is extended to time-varying graphs, and the results are applied to exponential family models, distributed logistic regression, and decentralized detection models.
