Posterior consistency of semi-supervised regression on graphs
From MaRDI portal
(Redirected from Publication:5157865)
Abstract: Graph-based semi-supervised regression (SSR) is the problem of estimating the value of a function on a weighted graph from its values (labels) on a small subset of the vertices. This paper is concerned with the consistency of SSR in the context of classification, in the setting where the labels have small noise and the underlying graph weighting is consistent with well-clustered nodes. We present a Bayesian formulation of SSR in which the weighted graph defines a Gaussian prior, using a graph Laplacian, and the labeled data defines a likelihood. We analyze the rate of contraction of the posterior measure around the ground truth in terms of parameters that quantify the small label error and inherent clustering in the graph. We obtain bounds on the rates of contraction and illustrate their sharpness through numerical experiments. The analysis also gives insight into the choice of hyperparameters that enter the definition of the prior.
Recommendations
- On the consistency of graph-based Bayesian semi-supervised learning and the scalability of sampling algorithms
- Large data and zero noise limits of graph-based semi-supervised learning algorithms
- Learning Theory
- Analysis of p-Laplacian regularization in semisupervised learning
- Properly-weighted graph Laplacian for semi-supervised learning
Cites work
- A variational approach to the consistency of spectral clustering
- An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach
- Analysis of support vector machines regression
- Asymptotic equivalence of nonparametric regression and white noise
- Asymptotic Statistics
- Bayesian inverse problems with Gaussian priors
- Consistency of semi-supervised learning algorithms on graphs: probit and one-hot methods
- Consistency of spectral clustering
- Consistency of Support Vector Machines and Other Regularized Kernel Classifiers
- Consistent Inversion of Noisy <scp>Non‐Abelian X‐Ray</scp> Transforms
- Convergence rates of posterior distributions.
- Diffuse interface models on graphs for classification of high dimensional data
- Error estimates for spectral convergence of the graph Laplacian on random geometric graphs toward the Laplace-Beltrami operator
- Estimating a smooth function on a large graph by Bayesian Laplacian regularisation
- Gaussian processes for machine learning.
- scientific article; zbMATH DE number 3599198 (Why is no real title available?)
- scientific article; zbMATH DE number 1332320 (Why is no real title available?)
- scientific article; zbMATH DE number 936298 (Why is no real title available?)
- Introduction to Bayesian Scientific Computing
- Kernel flows: from learning kernels from data into the abyss
- Learning semidefinite regularizers
- Learning Theory
- Manifold regularization: a geometric framework for learning from labeled and unlabeled examples
- MAP estimators and their consistency in Bayesian nonparametric inverse problems
- Mathematical foundations of infinite-dimensional statistical models
- Nonparametric Bayesian label prediction on a graph
- Nonparametric Bayesian label prediction on a large graph using truncated Laplacian regularization
- Nonstationary inverse problems and state estimation
- On the Bernstein-von Mises theorem with infinite-dimensional parameters
- On the consistency of Bayes estimates
- On the consistency of multiclass classification methods
- On the influence of the kernel on the consistency of support vector machines
- Posterior contraction rates for the Bayesian approach to linear ill-posed inverse problems
- Rates of contraction of posterior distributions based on Gaussian process priors
- Solving inverse problems using data-driven models
- Spectral partitioning works: planar graphs and finite element meshes
- The Rotation of Eigenvectors by a Perturbation. III
- Uncertainty quantification in graph-based classification of high dimensional data
Cited in
(6)- On the consistency of graph-based Bayesian semi-supervised learning and the scalability of sampling algorithms
- Consistency of semi-supervised learning algorithms on graphs: probit and one-hot methods
- Batch active learning for multispectral and hyperspectral image segmentation using similarity graphs
- Model change active learning in graph-based semi-supervised learning
- Graph Laplacian-based Bayesian multi-fidelity modeling
- Empirical stationary correlations for semi-supervised learning on graphs
This page was built for publication: Posterior consistency of semi-supervised regression on graphs
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5157865)
