Nonstationary Gaussian Process Discriminant Analysis With Variable Selection for High-Dimensional Functional Data
From MaRDI portal
Publication:6094087
DOI10.1080/10618600.2022.2098136arXiv2109.14171OpenAlexW3204897297WikidataQ114099323 ScholiaQ114099323MaRDI QIDQ6094087FDOQ6094087
Authors: Weichang Yu, Sara Wade, Howard D. Bondell, Lamiae Azizi 
Publication date: 9 October 2023
Published in: Journal of Computational and Graphical Statistics (Search for Journal in Brave)
Abstract: High-dimensional classification and feature selection tasks are ubiquitous with the recent advancement in data acquisition technology. In several application areas such as biology, genomics and proteomics, the data are often functional in their nature and exhibit a degree of roughness and non-stationarity. These structures pose additional challenges to commonly used methods that rely mainly on a two-stage approach performing variable selection and classification separately. We propose in this work a novel Gaussian process discriminant analysis (GPDA) that combines these steps in a unified framework. Our model is a two-layer non-stationary Gaussian process coupled with an Ising prior to identify differentially-distributed locations. Scalable inference is achieved via developing a variational scheme that exploits advances in the use of sparse inverse covariance matrices. We demonstrate the performance of our methodology on simulated datasets and two proteomics datasets: breast cancer and SARS-CoV-2. Our approach distinguishes itself by offering explainability as well as uncertainty quantification in addition to low computational cost, which are crucial to increase trust and social acceptance of data-driven tools.
Full work available at URL: https://arxiv.org/abs/2109.14171
Cites Work
- Two-group classification with high-dimensional correlated data: a factor model approach
- LIBLINEAR: a library for large linear classification
- Penalized classification using Fisher's linear discriminant
- Variational discriminant analysis with variable selection
- Bayesian functional linear regression with sparse step functions
- Support-vector networks
- Bayesian wavelet-based curve classification via discriminant analysis with Markov random tree priors
- Sparsity and Smoothness Via the Fused Lasso
- Multilayer Perceptron with Functional Inputs: an Inverse Regression Approach
- An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach
- High-dimensional classification using features annealed independence rules
- Some theory for Fisher's linear discriminant function, `naive Bayes', and some alternatives when there are many more variables than observations
- Penalized discriminant analysis
- Explaining variational approximations
- Title not available (Why is that?)
- How deep are deep Gaussian processes?
- Curves discrimination: a nonparametric functional approach
- Generalized Linear Models with Functional Predictors
- Scalable Gaussian Process Computations Using Hierarchical Matrices
- Penalized regression, standard errors, and Bayesian Lassos
- Accuracy and Stability of Numerical Algorithms
- Bayesian semiparametric joint models for functional predictors
- Time-Dependent Statistics of the Ising Model
- Functional linear discriminant analysis for irregularly sampled curves
- A Functional Data—Analytic Approach to Signal Discrimination
- Bayesian variable selection in structured high-dimensional covariate spaces with applications in genomics
- Bayesian nonparametric classification for spectroscopy data
- Spatial Bayesian variable selection and grouping for high-dimensional scalar-on-image regression
- Bayesian Wavelet Regression on Curves With Application to a Spectroscopic Calibration Problem
- Bayesian classification of multiclass functional data
- A Bayesian hierarchical model for classification with selection of functional predictors
- Variable selection and updating in model-based discriminant analysis for high dimensional data with food authenticity applications
- Model-free feature screening for ultrahigh dimensional discriminant analysis
- Methods for Scalar‐on‐Function Regression
- Regularized partial least squares with an application to NMR spectroscopy
- Diagonal Discriminant Analysis With Feature Selection for High-Dimensional Data
- Scalar-on-image regression via the soft-thresholded Gaussian process
- Introduction to statistical physics
- Deep state-space Gaussian processes
- Variational nonparametric discriminant analysis
- Posterior inference for sparse hierarchical non-stationary models
- Bayesian fusion estimation via \(t\) shrinkage
Cited In (1)
This page was built for publication: Nonstationary Gaussian Process Discriminant Analysis With Variable Selection for High-Dimensional Functional Data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6094087)
