A flexible model-free prediction-based framework for feature ranking

MaRDI QIDQ4999101zbMATH OpenFDO

Authors Jingyi Jessica Li, Yiling Elaine Chen, Xin Tong

Publication date 9 July 2021

Full work available at URL https://arxiv.org/abs/1903.05262, https://jmlr.csail.mit.edu/papers/v22/20-673.html

zbMATH Keywords

binary classificationsampling biasmodel-freeclassical and Neyman-Pearson paradigmsmarginal feature ranking

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• Prediction by Categorical Features: Generalization Properties and Application to Feature Ranking
• 10.1162/153244303322753733
• Model-free feature screening for ultrahigh dimensional classification
• scientific article; zbMATH DE number 1759575
• Ranking categorical features using generalization properties

Cites work

• A Neyman–Pearson Approach to Statistical Learning
• A plug-in approach to Neyman-Pearson classification
• An introduction to statistical learning. With applications in R
• Brownian distance covariance
• Fast learning rates for plug-in classifiers
• Gene selection for cancer classification using support vector machines
• Measuring mass concentrations and estimating density contour clusters -- An excess mass approach
• Multivariate plug-in bandwidth selection
• Neyman-Pearson classification under high-dimensional settings
• Neyman-Pearson classification: parametrics and sample size requirement
• Optimal rates for plug-in estimators of density level sets
• Oracle inequalities in empirical risk minimization and sparse recovery problems. École d'Été de Probabilités de Saint-Flour XXXVIII-2008.
• Smooth discrimination analysis

Cited in

(1)

• Feature ranking for multi-target regression

Describes a project that uses

Uses Software

• GWAS Catalog
• shap
• ElemStatLearn

This page was built for publication: A flexible model-free prediction-based framework for feature ranking

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4999101)