Random projections as regularizers: learning a linear discriminant from fewer observations than dimensions

DOI10.1007/s10994-014-5466-8zbMath1331.68183OpenAlexW2111662199WikidataQ58550247 ScholiaQ58550247MaRDI QIDQ2353006

Robert J. Durrant, Ata Kabán

Publication date: 7 July 2015

Published in: Machine Learning (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s10994-014-5466-8

zbMATH Keywords

learning theorylinear discriminant analysisensemble learningrandom projectionscompressed learning

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Related Items (9)

High-dimensional clustering via random projections ⋮ Unnamed Item ⋮ Unnamed Item ⋮ Increasing the accuracy of solving discrete ill-posed problems by the random projection method ⋮ Indefinite Proximity Learning: A Review ⋮ Sufficient ensemble size for random matrix theory-based handling of singular covariance matrices ⋮ Regularizing axis-aligned ensembles via data rotations that favor simpler learners ⋮ Correlations between random projections and the bivariate normal ⋮ Random projections as regularizers: learning a linear discriminant from fewer observations than dimensions

Uses Software

• LIBLINEAR
• rda
• ElemStatLearn
• CLOP
• SVM Toolbox

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Bagging predictors
• Regularized linear discriminant analysis and its application in microarrays
• A well-conditioned estimator for large-dimensional covariance matrices
• Database-friendly random projections: Johnson-Lindenstrauss with binary coins.
• Some theory for Fisher's linear discriminant function, `naive Bayes', and some alternatives when there are many more variables than observations
• Boosting the margin: a new explanation for the effectiveness of voting methods
• Empirical margin distributions and bounding the generalization error of combined classifiers
• Random projections as regularizers: learning a linear discriminant from fewer observations than dimensions
• An algorithmic theory of learning: Robust concepts and random projection
• Approximate nearest neighbors and the fast Johnson-Lindenstrauss transform
• On variants of the Johnson–Lindenstrauss lemma
• Matrix Analysis
• Expected classification error of the Fisher linear classifier with pseudo-inverse covariance matrix
• Comparison of Discrimination Methods for the Classification of Tumors Using Gene Expression Data
• Sure Independence Screening for Ultrahigh Dimensional Feature Space
• An elementary proof of a theorem of Johnson and Lindenstrauss
• On Consistency and Sparsity for Principal Components Analysis in High Dimensions
• A Random Matrix-Theoretic Approach to Handling Singular Covariance Estimates
• Singular vectors under random perturbation
• Random Matrix Theory and Wireless Communications
• The elements of statistical learning. Data mining, inference, and prediction
• Random forests

This page was built for publication: Random projections as regularizers: learning a linear discriminant from fewer observations than dimensions