Fixed-rank matrix factorizations and Riemannian low-rank optimization

DOI10.1007/s00180-013-0464-zzbMath1306.65107arXiv1209.0430OpenAlexW2042182457WikidataQ114231247 ScholiaQ114231247MaRDI QIDQ2259741

Bamdev Mishra, Gilles Meyer, Silvère Bonnabel, Rodolphe J. Sepulchre

Publication date: 5 March 2015

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

Full work available at URL: https://arxiv.org/abs/1209.0430

zbMATH Keywords

steepest descentlinear regressionlow-rank matrix completionRiemannian quotient geometryRiemannian trust-region

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Numerical solutions to overdetermined systems, pseudoinverses (65F20) Numerical mathematical programming methods (65K05) Learning and adaptive systems in artificial intelligence (68T05) Matrix completion problems (15A83)

Related Items

Guarantees of Riemannian optimization for low rank matrix completion, GNMR: A Provable One-Line Algorithm for Low Rank Matrix Recovery, Dynamically Orthogonal Numerical Schemes for Efficient Stochastic Advection and Lagrangian Transport, Low-rank tensor completion by Riemannian optimization, Finding stationary points on bounded-rank matrices: a geometric hurdle and a smooth remedy, Convergence Results for Projected Line-Search Methods on Varieties of Low-Rank Matrices Via Łojasiewicz Inequality, Dual dynamically orthogonal approximation of incompressible Navier Stokes equations with random boundary conditions, Low rank matrix completion by alternating steepest descent methods, A Riemannian Gradient Sampling Algorithm for Nonsmooth Optimization on Manifolds, A Geometric Approach to Dynamical Model Order Reduction, Geometric Methods on Low-Rank Matrix and Tensor Manifolds, Low-rank retractions: a survey and new results, Preconditioned Low-rank Riemannian Optimization for Linear Systems with Tensor Product Structure, Sparse matrices in data analysis, Two Newton methods on the manifold of fixed-rank matrices endowed with Riemannian quotient geometries, Riemannian Preconditioning, Riemannian gradient descent methods for graph-regularized matrix completion, Majorized proximal alternating imputation for regularized rank constrained matrix completion, Guarantees of Riemannian Optimization for Low Rank Matrix Recovery, Finding Low-Rank Solutions via Nonconvex Matrix Factorization, Efficiently and Provably, The Extrinsic Geometry of Dynamical Systems Tracking Nonlinear Matrix Projections, Accelerated Alternating Projections for Robust Principal Component Analysis, A Riemannian gossip approach to subspace learning on Grassmann manifold, Rank $2r$ Iterative Least Squares: Efficient Recovery of Ill-Conditioned Low Rank Matrices from Few Entries, A Three-Operator Splitting Algorithm for Nonconvex Sparsity Regularization

Uses Software

• PROPACK
• Manopt
• LMaFit
• softImpute
• GenRTR
• RTRMC
• Algorithm 432
• ADMiRA

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Solving a low-rank factorization model for matrix completion by a nonlinear successive over-relaxation algorithm
• Two Newton methods on the manifold of fixed-rank matrices endowed with Riemannian quotient geometries
• Low-rank matrix completion via preconditioned optimization on the Grassmann manifold
• Fast low-rank modifications of the thin singular value decomposition
• Grassmann algorithms for low rank approximation of matrices with missing values
• Positive definite matrices
• Exact matrix completion via convex optimization
• Low-Rank Matrix Completion by Riemannian Optimization
• Manopt, a Matlab toolbox for optimization on manifolds
• A Singular Value Thresholding Algorithm for Matrix Completion
• Full Rank Factorization of Matrices
• Riemannian Metric and Geometric Mean for Positive Semidefinite Matrices of Fixed Rank
• The Geometry of Algorithms with Orthogonality Constraints
• Subspace Evolution and Transfer (SET) for Low-Rank Matrix Completion
• Dimension Reduction and Coefficient Estimation in Multivariate Linear Regression
• LU factoring of non-invertible matrices
• A Geometric Approach to Low-Rank Matrix Completion
• Recovering Low-Rank Matrices From Few Coefficients in Any Basis
• ADMiRA: Atomic Decomposition for Minimum Rank Approximation
• Algorithm 432 [C2: Solution of the matrix equation AX + XB = C [F4]]
• Low-Rank Optimization with Trace Norm Penalty