Nonparametric Bayesian aggregation for massive data

zbMATH Open1441.62086arXiv1508.04175MaRDI QIDQ5214232FDOQ5214232

Authors: Zuofeng Shang, Botao Hao, Guang Cheng 

Publication date: 7 February 2020

Abstract: We develop a set of scalable Bayesian inference procedures for a general class of nonparametric regression models. Specifically, nonparametric Bayesian inferences are separately performed on each subset randomly split from a massive dataset, and then the obtained local results are aggregated into global counterparts. This aggregation step is explicit without involving any additional computation cost. By a careful partition, we show that our aggregated inference results obtain an oracle rule in the sense that they are equivalent to those obtained directly from the entire data (which are computationally prohibitive). For example, an aggregated credible ball achieves desirable credibility level and also frequentist coverage while possessing the same radius as the oracle ball.

Full work available at URL: https://arxiv.org/abs/1508.04175

Recommendations

• Nonparametric Bayesian data analysis
• Nonparametric Bayesian inference in applications
• A Method for Combining Inference Across Related Nonparametric Bayesian Models
• Scalable Bayes via barycenter in Wasserstein space
• Distributed Bayesian posterior voting strategy for massive data

zbMATH Keywords

nonparametric Bayesian inferenceGaussian process priorlinear functionaldivide-and-conquercredible region

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Statistical aspects of big data and data science (62R07) Gaussian processes (60G15) Nonparametric tolerance and confidence regions (62G15)

Cites Work

• Introduction to empirical processes and semiparametric inference
• A central limit theorem for generalized quadratic forms
• Title not available (Why is that?)
• Convergence rates of posterior distributions.
• Title not available (Why is that?)
• Nonparametric Bernstein-von Mises theorems in Gaussian white noise
• Title not available (Why is that?)
• On the Bernstein-von Mises phenomenon for nonparametric Bayes procedures
• Natural exponential families with quadratic variance functions: Statistical theory
• Prevalence: a translation-invariant “almost every” on infinite-dimensional spaces
• Title not available (Why is that?)
• Bernstein-von Mises theorem for linear functionals of the density
• Rates of contraction of posterior distributions based on Gaussian process priors
• On the lower tail of Gaussian seminorms
• Transformations of Wiener integrals under translations
• Local and global asymptotic inference in smoothing spline models
• Divide and conquer kernel ridge regression: a distributed algorithm with minimax optimal rates
• A partially linear framework for massive heterogeneous data
• The Gaussian measure of shifted balls
• Optimum bounds for the distributions of martingales in Banach spaces
• Pushing the Limits of Contemporary Statistics: Contributions in Honor of Jayanta K. Ghosh
• A Scalable Bootstrap for Massive Data
• A Gaussian correlation inequality and its applications to small ball probabilities
• Robust and scalable Bayes via a median of subset posterior measures
• Gaussian approximation of general non-parametric posterior distributions
• Simple, scalable and accurate posterior interval estimation
• Scalable Bayes via barycenter in Wasserstein space
• Nonparametric Bayesian aggregation for massive data
• Adaptive distributed methods under communication constraints
• An asymptotic analysis of distributed nonparametric methods

Cited In (15)

• Nonparametric Bayesian aggregation for massive data
• Distributed Bayesian inference in massive spatial data
• Distributed adaptive nearest neighbor classifier: algorithm and theory
• Fully Bayesian aggregation
• Scalable Bayesian nonparametric regression via a Plackett-Luce model for conditional ranks
• Smoothing Splines Approximation Using Hilbert Curve Basis Selection
• Robust Bayesian inference for big data: combining sensor-based records with traditional survey data
• Nonparametric distributed learning under general designs
• A distributed multiple sample testing for massive data
• Fast algorithms for nonparametric population modeling of large data sets
• Partitioned Approach for High-dimensional Confidence Intervals with Large Split Sizes
• Distributed Bayesian posterior voting strategy for massive data
• Emerging directions in Bayesian computation
• Scalable Bayes via barycenter in Wasserstein space
• Towards Gaussian Bayesian network fusion

This page was built for publication: Nonparametric Bayesian aggregation for massive data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5214232)