A global two-stage algorithm for non-convex penalized high-dimensional linear regression problems

DOI10.1007/s00180-022-01249-warXiv2111.11801OpenAlexW3215217380MaRDI QIDQ6177008

Zhou Yu, Min Liu, Peili Li

Publication date: 29 August 2023

Published in: Computational Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2111.11801

zbMATH Keywords

difference of convex functionsglobal convergencehigh-dimensional linear regressiontwo-stage algorithmprimal dual active set with continuation algorithm

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08)

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Coordinate descent algorithms for nonconvex penalized regression, with applications to biological feature selection
• Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers
• Nearly unbiased variable selection under minimax concave penalty
• DC approximation approaches for sparse optimization
• One-step sparse estimates in nonconcave penalized likelihood models
• Calculus of the exponent of Kurdyka-Łojasiewicz inequality and its applications to linear convergence of first-order methods
• Least angle regression. (With discussion)
• A unified primal dual active set algorithm for nonconvex sparse recovery
• Calibrating nonconvex penalized regression in ultra-high dimension
• Coordinate descent algorithms for lasso penalized regression
• High-dimensional graphs and variable selection with the Lasso
• Variable selection using MM algorithms
• Atomic Decomposition by Basis Pursuit
• Proximal Newton-Type Methods for Minimizing Composite Functions
• Computing B-Stationary Points of Nonsmooth DC Programs
• Proximity algorithms for image models: denoising
• SparseNet: Coordinate Descent With Nonconvex Penalties
• Adapting to Unknown Smoothness via Wavelet Shrinkage
• Decoding by Linear Programming
• The Primal-Dual Active Set Strategy as a Semismooth Newton Method
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
• Composite Difference-Max Programs for Modern Statistical Estimation Problems
• The Convergence Guarantees of a Non-Convex Approach for Sparse Recovery
• A Primal Dual Active Set Algorithm With Continuation for Compressed Sensing
• A Highly Efficient Semismooth Newton Augmented Lagrangian Method for Solving Lasso Problems
• A Statistical View of Some Chemometrics Regression Tools
• Sparse Approximate Solutions to Linear Systems
• Sequential Lasso Cum EBIC for Feature Selection With Ultra-High Dimensional Feature Space
• Difference-of-Convex Learning: Directional Stationarity, Optimality, and Sparsity

This page was built for publication: A global two-stage algorithm for non-convex penalized high-dimensional linear regression problems