Matrix completion and low-rank SVD via fast alternating least squares
From MaRDI portal
Publication:97013
DOI10.48550/ARXIV.1410.2596zbMATH Open1352.65117arXiv1410.2596MaRDI QIDQ97013FDOQ97013
Authors: Rahul Mazumder, Jason D. Lee, Reza Zadeh, Trevor Hastie 
Publication date: 9 October 2014
Published in: Journal of Machine Learning Research (JMLR) (Search for Journal in Brave)
Abstract: The matrix-completion problem has attracted a lot of attention, largely as a result of the celebrated Netflix competition. Two popular approaches for solving the problem are nuclear-norm-regularized matrix approximation (Candes and Tao, 2009, Mazumder, Hastie and Tibshirani, 2010), and maximum-margin matrix factorization (Srebro, Rennie and Jaakkola, 2005). These two procedures are in some cases solving equivalent problems, but with quite different algorithms. In this article we bring the two approaches together, leading to an efficient algorithm for large matrix factorization and completion that outperforms both of these. We develop a software package "softImpute" in R for implementing our approaches, and a distributed version for very large matrices using the "Spark" cluster programming environment.
Full work available at URL: https://arxiv.org/abs/1410.2596
Recommendations
- Spectral regularization algorithms for learning large incomplete matrices
- Solving a low-rank factorization model for matrix completion by a nonlinear successive over-relaxation algorithm
- A Singular Value Thresholding Algorithm for Matrix Completion
- Low-rank matrix completion using alternating minimization
- Fixed point and Bregman iterative methods for matrix rank minimization
Estimation in multivariate analysis (62H12) Matrix completion problems (15A83) Numerical computation of eigenvalues and eigenvectors of matrices (65F15)
Cited In (73)
- Variable selection for high-dimensional incomplete data
- Imputed mean tensor regression for near-sited spatial temporal data
- Imputed quantile tensor regression for near-sited spatial-temporal data
- Column \(\ell_{2,0}\)-norm regularized factorization model of low-rank matrix recovery and its computation
- Matrix completion from a computational statistics perspective
- Large factor model estimation by nuclear norm plus \(\ell_1\) norm penalization
- A data-adaptive dimension reduction for functional data via penalized low-rank approximation
- Convergence analysis of the rank-restricted soft SVD algorithm
- Collaborative Filtering With Awareness of Social Networks
- Reexamining low rank matrix factorization for trace norm regularization
- Core-elements for large-scale least squares estimation
- Approximate matrix completion based on cavity method
- Scenario-based quantile connectedness of the U.S. interbank liquidity risk network
- Ridge Regularization: An Essential Concept in Data Science
- Stable als approximation in the TT-format for rank-adaptive tensor completion
- A matrix completion algorithm using randomized SVD
- Matrix Completion Methods for Causal Panel Data Models
- Setting up alternating least squares and iterative majorization algorithms for solving various matrix optimization problems
- A singular value thresholding with diagonal-update algorithm for low-rank matrix completion
- Outlier detection in networks with missing links
- A very fast algorithm for matrix factorization
- Flexible low-rank statistical modeling with missing data and side information
- Spectral regularization algorithms for learning large incomplete matrices
- Smooth over-parameterized solvers for non-smooth structured optimization
- Rank regularized estimation of approximate factor models
- Distributed matrix completion and robust factorization
- Recovery of simultaneous low rank and two-way sparse coefficient matrices, a nonconvex approach
- Implicit regularization in nonconvex statistical estimation: gradient descent converges linearly for phase retrieval, matrix completion, and blind deconvolution
- Low-rank traffic matrix completion with marginal information
- Missing value imputation in multivariate time series with end-to-end generative adversarial networks
- On a Problem of Weighted Low-Rank Approximation of Matrices
- Low-rank matrix completion using alternating minimization
- Matrix completion under interval uncertainty
- Accurate low-rank approximations via a few iterations of alternating least squares
- A nonlinear matrix decomposition for mining the zeros of sparse data
- Sparse principal component analysis with missing observations
- Low-rank factorization for rank minimization with nonconvex regularizers
- An Alternating Method for Cardinality-Constrained Optimization: A Computational Study for the Best Subset Selection and Sparse Portfolio Problems
- Riemannian gradient descent methods for graph-regularized matrix completion
- Provable accelerated gradient method for nonconvex low rank optimization
- A Trust-region Method for Nonsmooth Nonconvex Optimization
- Inductive matrix completion with feature selection
- Matrix completion with noisy entries and outliers
- Title not available (Why is that?)
- Imputation and low-rank estimation with missing not at random data
- Inference for low-rank models
- Active matrix factorization for surveys
- Matrix completion discriminant analysis
- Imputation of Mixed Data With Multilevel Singular Value Decomposition
- Title not available (Why is that?)
- \textsc{OCam}: out-of-core coordinate descent algorithm for matrix completion
- softImpute
- Greedy low-rank algorithm for spatial connectome regression
- Matrix completion methods for the total electron content video reconstruction
- Error bound of critical points and KL property of exponent 1/2 for squared F-norm regularized factorization
- rsparse
- Title not available (Why is that?)
- Title not available (Why is that?)
- The power of subsampling in submodular maximization
- Orthogonal rank-one matrix pursuit for low rank matrix completion
- Title not available (Why is that?)
- Title not available (Why is that?)
- Smooth neighborhood recommender systems
- Intelligent Initialization and Adaptive Thresholding for Iterative Matrix Completion: Some Statistical and Algorithmic Theory forAdaptive-Impute
- Majorized proximal alternating imputation for regularized rank constrained matrix completion
- Matrix completion with nonconvex regularization: spectral operators and scalable algorithms
- Numerical comparisons between Bayesian and frequentist low-rank matrix completion: estimation accuracy and uncertainty quantification
- Main effects and interactions in mixed and incomplete data frames
- Empirical Bayes matrix completion
- Automatic Transformation and Integration to Improve Visualization and Discovery of Latent Effects in Imaging Data
- Fast methods for denoising matrix completion formulations, with applications to robust seismic data interpolation
- Nonparametric Imputation by Data Depth
- A fast regression via SVD and marginalization
Uses Software
This page was built for publication: Matrix completion and low-rank SVD via fast alternating least squares
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q97013)
