A reduced-rank approach to predicting multiple binary responses through machine learning

DOI10.1007/S11222-023-10314-3zbMATH Open1523.62025arXiv2306.05829OpenAlexW4387618032MaRDI QIDQ6089192FDOQ6089192

Authors: The Tien Mai 

Publication date: 17 November 2023

Published in: Statistics and Computing (Search for Journal in Brave)

Abstract: This paper investigates the problem of simultaneously predicting multiple binary responses by utilizing a shared set of covariates. Our approach incorporates machine learning techniques for binary classification, without making assumptions about the underlying observations. Instead, our focus lies on a group of predictors, aiming to identify the one that minimizes prediction error. Unlike previous studies that primarily address estimation error, we directly analyze the prediction error of our method using PAC-Bayesian bounds techniques. In this paper, we introduce a pseudo-Bayesian approach capable of handling incomplete response data. Our strategy is efficiently implemented using the Langevin Monte Carlo method. Through simulation studies and a practical application using real data, we demonstrate the effectiveness of our proposed method, producing comparable or sometimes superior results compared to the current state-of-the-art method.

Full work available at URL: https://arxiv.org/abs/2306.05829

Recommendations

• Bayesian classification based on multivariate binary data
• High-dimensional pseudo-logistic regression and classification with applications to gene expression data
• Leveraging mixed and incomplete outcomes via reduced-rank modeling
• Improving multilabel classification via heterogeneous ensemble methods
• Clustering of multivariate binary data with dimension reduction via \(L_{1}\)-regularized likelihood maximization

zbMATH Keywords

missing dataPAC-Bayesian inequalitiesbinary responsesLangevin Monte Carlolow-rank predictors

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Bayesian inference (62F15) Learning and adaptive systems in artificial intelligence (68T05) Estimation in multivariate analysis (62H12)

Cites Work

• Reduced rank regression via adaptive nuclear norm penalization
• Leveraging mixed and incomplete outcomes via reduced-rank modeling
• PAC-Bayesian analysis of co-clustering and beyond
• Title not available (Why is that?)
• Modern Multivariate Statistical Techniques
• Reduced-rank regression for the multivariate linear model
• Robust reduced-rank regression
• Optimal selection of reduced rank estimators of high-dimensional matrices
• Estimating Linear Restrictions on Regression Coefficients for Multivariate Normal Distributions
• Langevin diffusions and Metropolis-Hastings algorithms
• 1-bit matrix completion: PAC-Bayesian analysis of a variational approximation
• A Bayesian approach for noisy matrix completion: optimal rate under general sampling distribution
• Bayesian methods for low-rank matrix estimation: short survey and theoretical study
• Optimal Scaling of Discrete Approximations to Langevin Diffusions
• Title not available (Why is that?)
• Fast Low-Rank Bayesian Matrix Completion With Hierarchical Gaussian Prior Models
• Title not available (Why is that?)
• Pseudo-Bayesian quantum tomography with rank-adaptation
• Title not available (Why is that?)
• Smooth discrimination analysis
• Statistical behavior and consistency of classification methods based on convex risk minimization.
• Inconsistency of Bayesian inference for misspecified linear models, and a proposal for repairing it
• A General Framework for Updating Belief Distributions
• Ranking and empirical minimization of \(U\)-statistics
• PAC-Bayesian Inequalities for Martingales
• Priors, posteriors and Bayes factors for a Bayesian analysis of cointegration
• Bayesian reduced rank regression in econometrics
• Statistical learning theory and stochastic optimization. Ecole d'Eté de Probabilitiés de Saint-Flour XXXI -- 2001.
• Upper bounds and aggregation in bipartite ranking
• Aggregation by exponential weighting, sharp PAC-Bayesian bounds and sparsity
• Sparse regression learning by aggregation and Langevin Monte-Carlo
• A PAC-Bayesian margin bound for linear classifiers
• High-dimensional Bayesian inference via the unadjusted Langevin algorithm
• On the properties of variational approximations of Gibbs posteriors
• Title not available (Why is that?)
• Bayesian assessment of dimensionality in reduced rank regression
• Bayesian sparse reduced rank multivariate regression
• Exponential weights in multivariate regression and a low-rankness favoring prior
• An Introduction to Envelopes
• Calibrating general posterior credible regions
• General Bayesian updating and the loss-likelihood bootstrap
• Risk bounds for the majority vote: from a PAC-Bayesian analysis to a learning algorithm
• General Bayesian Loss Function Selection and the use of Improper Models
• Bayesian sparse multiple regression for simultaneous rank reduction and variable selection
• A comparison of learning rate selection methods in generalized Bayesian inference
• Low-Rank Regression Models for Multiple Binary Responses and their Applications to Cancer Cell-Line Encyclopedia Data
• Adaptation of the tuning parameter in general Bayesian inference with robust divergence

This page was built for publication: A reduced-rank approach to predicting multiple binary responses through machine learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6089192)