High-dimensional model recovery from random sketched data by exploring intrinsic sparsity

DOI10.1007/S10994-019-05865-4zbMath1472.68165OpenAlexW3000071498WikidataQ126399604 ScholiaQ126399604MaRDI QIDQ782446

Qihang Lin, Shenghuo Zhu, Rong Jin, Tianbao Yang, Li-jun Zhang

Publication date: 27 July 2020

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

Full work available at URL: https://doi.org/10.1007/s10994-019-05865-4

zbMATH Keywords

classificationregressionsparsityhigh dimensionJL-transformrandomized reduction

Mathematics Subject Classification ID

Ridge regression; shrinkage estimators (Lasso) (62J07) Classification and discrimination; cluster analysis (statistical aspects) (62H30) General nonlinear regression (62J02) Learning and adaptive systems in artificial intelligence (68T05)

Uses Software

• COFFIN
• RCV1
• Saga

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Finding structure with randomness: Probabilistic algorithms for constructing approximate matrix decompositions
• CUR matrix decompositions for improved data analysis
• Oracle inequalities in empirical risk minimization and sparse recovery problems. École d'Été de Probabilités de Saint-Flour XXXVIII-2008.
• Faster least squares approximation
• Learning to decode cognitive states from brain images
• ``Preconditioning for feature selection and regression in high-dimensional problems
• Database-friendly random projections: Johnson-Lindenstrauss with binary coins.
• Preconditioning the Lasso for sign consistency
• Simultaneous analysis of Lasso and Dantzig selector
• Image classification with the Fisher vector: theory and practice
• The Dantzig selector: statistical estimation when \(p\) is much larger than \(n\). (With discussions and rejoinder).
• Iterative Hessian sketch: Fast and accurate solution approximation for constrained least-squares
• One-Bit Compressed Sensing by Linear Programming
• A Proximal-Gradient Homotopy Method for the Sparse Least-Squares Problem
• Improved Matrix Algorithms via the Subsampled Randomized Hadamard Transform
• A sparse Johnson
• Approximate nearest neighbors and the fast Johnson-Lindenstrauss transform
• Randomized Sketches of Convex Programs With Sharp Guarantees
• Random Projections for Classification: A Recovery Approach
• IMPROVED ANALYSIS OF THE SUBSAMPLED RANDOMIZED HADAMARD TRANSFORM
• Randomized Algorithms for Matrices and Data
• Sparser Johnson-Lindenstrauss Transforms
• Extensions of Lipschitz mappings into a Hilbert space
• Sampling algorithms for l₂ regression and applications
• Relative-Error $CUR$ Matrix Decompositions
• An elementary proof of a theorem of Johnson and Lindenstrauss
• Information-Theoretic Limits on Sparsity Recovery in the High-Dimensional and Noisy Setting
• The Fast Johnson–Lindenstrauss Transform and Approximate Nearest Neighbors
• A Proximal Stochastic Gradient Method with Progressive Variance Reduction
• Regularization and Variable Selection Via the Elastic Net
• Stochastic Dual Coordinate Ascent Methods for Regularized Loss Minimization
• Algorithmic Learning Theory
• Gene selection for cancer classification using support vector machines
• Accelerated proximal stochastic dual coordinate ascent for regularized loss minimization

This page was built for publication: High-dimensional model recovery from random sketched data by exploring intrinsic sparsity