Uncertainty quantification in graph-based classification of high dimensional data
From MaRDI portal
Abstract: Classification of high dimensional data finds wide-ranging applications. In many of these applications equipping the resulting classification with a measure of uncertainty may be as important as the classification itself. In this paper we introduce, develop algorithms for, and investigate the properties of, a variety of Bayesian models for the task of binary classification; via the posterior distribution on the classification labels, these methods automatically give measures of uncertainty. The methods are all based around the graph formulation of semi-supervised learning. We provide a unified framework which brings together a variety of methods which have been introduced in different communities within the mathematical sciences. We study probit classification in the graph-based setting, generalize the level-set method for Bayesian inverse problems to the classification setting, and generalize the Ginzburg-Landau optimization-based classifier to a Bayesian setting; we also show that the probit and level set approaches are natural relaxations of the harmonic function approach introduced in [Zhu et al 2003]. We introduce efficient numerical methods, suited to large data-sets, for both MCMC-based sampling as well as gradient-based MAP estimation. Through numerical experiments we study classification accuracy and uncertainty quantification for our models; these experiments showcase a suite of datasets commonly used to evaluate graph-based semi-supervised learning algorithms.
Recommendations
- Graph-based optimization approaches for machine learning, uncertainty quantification and networks
- Diffuse Interface Models on Graphs for Classification of High Dimensional Data
- Nonparametric Bayesian label prediction on a graph
- On the consistency of graph-based Bayesian semi-supervised learning and the scalability of sampling algorithms
- Diffuse interface methods for multiclass segmentation of high-dimensional data
Cites work
- scientific article; zbMATH DE number 45848 (Why is no real title available?)
- scientific article; zbMATH DE number 1522714 (Why is no real title available?)
- scientific article; zbMATH DE number 762973 (Why is no real title available?)
- scientific article; zbMATH DE number 6253975 (Why is no real title available?)
- scientific article; zbMATH DE number 964896 (Why is no real title available?)
- A Bayesian level set method for geometric inverse problems
- An MBO scheme on graphs for classification and image processing
- Convergence of the graph Allen-Cahn scheme
- Diffuse Interface Models on Graphs for Classification of High Dimensional Data
- Diffuse interface methods for multiclass segmentation of high-dimensional data
- Diffuse interface models on graphs for classification of high dimensional data
- Hierarchical Bayesian level set inversion
- Introduction to uncertainty quantification
- Learning Theory
- MCMC METHODS FOR DIFFUSION BRIDGES
- MCMC methods for functions: modifying old algorithms to make them faster
- Manifold regularization: a geometric framework for learning from labeled and unlabeled examples
- Minimal Dirichlet energy partitions for graphs
- Multi-parameter regularization and its numerical realization
- Nonparametric Bayesian label prediction on a graph
- Numerical methods for stochastic computations. A spectral method approach.
- On the brittleness of Bayesian inference
- Optimal scalings for local Metropolis-Hastings chains on nonproduct targets in high dimensions
- Optimal uncertainty quantification
- Pattern recognition and machine learning.
- Uncertainty quantification. Theory, implementation, and applications
- Wavelets on graphs via spectral graph theory
- Weak convergence and optimal scaling of random walk Metropolis algorithms
- \(\Gamma\)-convergence of graph Ginzburg-Landau functionals
Cited in
(38)- Classification and image processing with a semi‐discrete scheme for fidelity forced Allen–Cahn on graphs
- Improved spectral convergence rates for graph Laplacians on \(\varepsilon \)-graphs and \(k\)-NN graphs
- Multilevel approximation of Gaussian random fields: fast simulation
- Properly-weighted graph Laplacian for semi-supervised learning
- Lipschitz regularity of graph Laplacians on random data clouds
- scientific article; zbMATH DE number 7306882 (Why is no real title available?)
- Kernel Methods for Bayesian Elliptic Inverse Problems on Manifolds
- Graph-theoretic algorithms for Kolmogorov operators: approximating solutions and their gradients in elliptic and parabolic problems on manifolds
- On the well-posedness of Bayesian inverse problems
- Batch active learning for multispectral and hyperspectral image segmentation using similarity graphs
- Model change active learning in graph-based semi-supervised learning
- Analysis of p-Laplacian regularization in semisupervised learning
- Gradient flows and randomised thresholding: sparse inversion and classification
- Minimax lower bounds for function estimation on graphs
- Stability of Gibbs posteriors from the Wasserstein loss for Bayesian full waveform inversion
- Data Based Construction of Kernels for Classification
- Ensemble Kalman inversion: a derivative-free technique for machine learning tasks
- Semi-supervised learning with summary statistics
- Stochastic block models are a discrete surface tension
- Graph Laplacian-based Bayesian multi-fidelity modeling
- A maximum principle argument for the uniform convergence of graph Laplacian regressors
- Continuum limits of posteriors in graph Bayesian inverse problems
- Local regularization of noisy point clouds: improved global geometric estimates and data analysis
- Gaussian measures conditioned on nonlinear observations: consistency, MAP estimators, and simulation
- Variational limits of \(k\)-NN graph-based functionals on data clouds
- Graph-based optimization approaches for machine learning, uncertainty quantification and networks
- Bayesian Inverse Problems Are Usually Well-Posed
- Gaussian process landmarking on manifolds
- Posterior consistency of semi-supervised regression on graphs
- Graph-based prior and forward models for inverse problems on manifolds with boundaries
- Spectral analysis of weighted Laplacians arising in data clustering
- On the Laplace equation on bounded subanalytic manifolds
- Data-driven prediction of multistable systems from sparse measurements
- The SPDE approach to Matérn fields: graph representations
- Spectral gaps and error estimates for infinite-dimensional Metropolis-Hastings with non-Gaussian priors
- The Bayesian update: variational formulations and gradient flows
- Nonparametric Bayesian label prediction on a large graph using truncated Laplacian regularization
- On the consistency of graph-based Bayesian semi-supervised learning and the scalability of sampling algorithms
This page was built for publication: Uncertainty quantification in graph-based classification of high dimensional data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3176234)
