Gibbs posterior for variable selection in high-dimensional classification and data mining
From MaRDI portal
Abstract: In the popular approach of "Bayesian variable selection" (BVS), one uses prior and posterior distributions to select a subset of candidate variables to enter the model. A completely new direction will be considered here to study BVS with a Gibbs posterior originating in statistical mechanics. The Gibbs posterior is constructed from a risk function of practical interest (such as the classification error) and aims at minimizing a risk function without modeling the data probabilistically. This can improve the performance over the usual Bayesian approach, which depends on a probability model which may be misspecified. Conditions will be provided to achieve good risk performance, even in the presence of high dimensionality, when the number of candidate variables "" can be much larger than the sample size "." In addition, we develop a convenient Markov chain Monte Carlo algorithm to implement BVS with the Gibbs posterior.
Recommendations
- Bayesian variable selection with shrinking and diffusing priors
- On the Consistency of Bayesian Variable Selection for High Dimensional Binary Regression and Classification
- Bayesian variable selection for high dimensional generalized linear models: convergence rates of the fitted densities
- Skinny Gibbs: a consistent and scalable Gibbs sampler for model selection
- Sparse Bayesian variable selection in kernel probit model for analyzing high-dimensional data
Cites work
- A Smoothed Maximum Score Estimator for the Binary Response Model
- Bayesian variable selection for high dimensional generalized linear models: convergence rates of the fitted densities
- Best subset selection, persistence in high-dimensional statistical learning and optimization under l₁ constraint
- Discussion on boosting papers.
- From -entropy to KL-entropy: analysis of minimum information complexity density estima\-tion
- scientific article; zbMATH DE number 1034042 (Why is no real title available?)
- scientific article; zbMATH DE number 893887 (Why is no real title available?)
- scientific article; zbMATH DE number 3249515 (Why is no real title available?)
- Information-theoretic upper and lower bounds for statistical estimation
- Misspecification in infinite-dimensional Bayesian statistics
- Nonparametric regression using Bayesian variable selection
- Sparse graphical models for exploring gene expression data
- Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
- The Calculation of Posterior Distributions by Data Augmentation
- The choice of variables in multivariate regression: a non-conjugate Bayesian decision theory approach
- Theory & Methods: Bayesian variable selection in logistic regression: predicting company earnings direction
- Tools for statistical inference. Methods for the exploration of posterior distributions and likelihood functions.
Cited in
(51)- `Purposely misspecified' posterior inference on the volatility of a jump diffusion process
- Bayesian fractional posteriors
- Robust Bayes-like estimation: rho-Bayes estimation
- Gibbs posterior inference on multivariate quantiles
- Bayes posterior convergence for loss functions via almost additive thermodynamic formalism
- Gibbs posterior convergence and the thermodynamic formalism
- Model-free posterior inference on the area under the receiver operating characteristic curve
- Regularization of Bayesian quasi-likelihoods constructed from complex estimating functions
- \(\alpha\)-variational inference with statistical guarantees
- Robust and rate-optimal Gibbs posterior inference on the boundary of a noisy image
- Bayesian inference on volatility in the presence of infinite jump activity and microstructure noise
- On general Bayesian inference using loss functions
- Quasi-Bayesian analysis of nonparametric instrumental variables models
- Bayesian model robustness via disparities
- Robust Bayes estimation using the density power divergence
- Gibbs posterior concentration rates under sub-exponential type losses
- General inequalities for Gibbs posterior with nonadditive empirical risk
- On the properties of variational approximations of Gibbs posteriors
- Predicting Panel Data Binary Choice with the Gibbs Posterior
- On extensions of Hoeffding's inequality for panel data
- Posterior consistency of nonparametric conditional moment restricted models
- Asymptotically minimax empirical Bayes estimation of a sparse normal mean vector
- High-dimensional Bayesian inference in nonparametric additive models
- Risk minimization for time series binary choice with variable selection
- GemBag: group estimation of multiple Bayesian graphical models
- General Robust Bayes Pseudo-Posteriors: Exponential Convergence Results with Applications
- Ensemble Kalman methods for high-dimensional hierarchical dynamic space-time models
- Asymptotic normality, concentration, and coverage of generalized posteriors
- Gibbs posterior inference on value-at-risk
- On oracle property and asymptotic validity of Bayesian generalized method of moments
- Generalized Bayes approach to inverse problems with model misspecification
- Generalized Bayes Quantification Learning under Dataset Shift
- Robust estimation in controlled branching processes: Bayesian estimators via disparities
- User-friendly Introduction to PAC-Bayes Bounds
- From robust tests to Bayes-like posterior distributions
- Joint production in stochastic non-parametric envelopment of data with firm-specific directions
- Direct Gibbs posterior inference on risk minimizers: construction, concentration, and calibration
- Adaptive variable selection for sequential prediction in multivariate dynamic models
- Approximating Bayes in the 21st century
- Bayesian Multi-Task Variable Selection with an Application to Differential DAG Analysis
- Probabilistic contrastive dimension reduction for case-control study data
- Forecasting the risk of cryptocurrencies: comparison and combination of GARCH and stochastic volatility models
- Bayesian inference of multiple Ising models for heterogeneous public opinion survey networks
- Robust Bayesian model averaging for linear regression models with heavy-tailed errors
- Repulsion, chaos, and equilibrium in mixture models
- Loss-Based Variational Bayes Prediction
- Laplace and saddlepoint approximations in high dimensions
- Characterizing asymptotic dependence between a satellite precipitation product and station data in the Northern US Rocky Mountains via the tail dependence regression framework with a Gibbs posterior inference approach
- Generalized geographically weighted regression model within a modularized Bayesian framework
- Sequential Gibbs posteriors with applications to principal component analysis
- A note on some algorithms for the Gibbs posterior
This page was built for publication: Gibbs posterior for variable selection in high-dimensional classification and data mining
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q955139)
