A New Block-Splitting Method for Nonnegative Low-Rank Matrix Approximation

Nonnegative low-rank matrix approximation is an important technique in data analysis for extracting meaningful patterns from high-dimensional nonnegative data. This nonnegative low-rank approximation problem is studied and a new block-splitting method is developed in this paper. This new method enforces the low-rank constraint by utilizing QR decomposition and adopts a semismooth Newton method to address the related convex subproblems efficiently through the dual formulation of the nonnegative low-rank matrix approximation problem. Theoretical analysis confirms the convergence of the new method. Several real datasets are used to demonstrate the efficiency of the proposed method.

• Publication year

  2026

• Venue

  IEEE Transactions on Knowledge and Data Engineering

• Publication date

  2026-03-01

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.1109/TKDE.2026.3655780
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Convergence of a Fast Hierarchical Alternating Least Squares Algorithm for Nonnegative Matrix Factorization

  2024cited by this paper
• Accelerated SVD-based initialization for nonnegative matrix factorization

  2024cited by this paper
• Explainable machine learning approach for cancer prediction through binarilization of RNA sequencing data

  2024cited by this paper
• Nonnegative tensor train for the multicomponent Smoluchowski equation

  2024cited by this paper
• Study of performance of low-rank nonnegative tensor factorization methods

  2023cited by this paper
• Quasioptimal alternating projections and their use in low-rank approximation of matrices and tensors

  2023cited by this paper
• Sketching for a low-rank nonnegative matrix approximation: Numerical study

  2022influential reference
• A Progressive Hierarchical Alternating Least Squares Method for Symmetric Nonnegative Matrix Factorization

  2022cited by this paper
• Application of non-negative matrix factorization in oncology: one approach for establishing precision medicine

  2022cited by this paper
• An Approximate Augmented Lagrangian Method for Nonnegative Low-Rank Matrix Approximation

  2021influential reference
• Tangent Space Based Alternating Projections for Nonnegative Low Rank Matrix Approximation

  2020influential reference
• Nonnegative low rank matrix approximation for nonnegative matrices

  2019influential reference
• Methods for Nonnegative Matrix Factorization Based on Low-Rank Cross Approximations

  2019cited by this paper
• Improved SVD-based Initialization for Nonnegative Matrix Factorization using Low-Rank Correction

  2018cited by this paper
• Rank-One NMF-Based Initialization for NMF and Relative Error Bounds Under a Geometric Assumption

  2016cited by this paper
• Introduction to Nonlinear Optimization - Theory, Algorithms, and Applications with MATLAB

  2014influential reference
• New SVD based initialization strategy for non-negative matrix factorization

  2014cited by this paper
• Algorithms for nonnegative matrix and tensor factorizations: a unified view based on block coordinate descent framework

  2013cited by this paper
• Convex optimization

  2010cited by this paper
• On the Use of the Adjusted Rand Index as a Metric for Evaluating Supervised Classification

  2009cited by this paper
• Fast Local Algorithms for Large Scale Nonnegative Matrix and Tensor Factorizations

  2009cited by this paper
• SVD based initialization: A head start for nonnegative matrix factorization

  2008cited by this paper
• Projected Gradient Methods for Nonnegative Matrix Factorization

  2007cited by this paper
• Hierarchical ALS Algorithms for Nonnegative Matrix and 3D Tensor Factorization

  2007cited by this paper
• A Dual Optimization Approach to Inverse Quadratic Eigenvalue Problems with Partial Eigenstructure

  2007cited by this paper
• Computing the Nearest Doubly Stochastic Matrix with A Prescribed Entry

  2007cited by this paper
• The Relationships Among Various Nonnegative Matrix Factorization Methods for Clustering

  2006cited by this paper
• Semismooth Matrix-Valued Functions

  2002cited by this paper
• Cluster Ensembles --- A Knowledge Reuse Framework for Combining Multiple Partitions

  2002cited by this paper
• Non-negative Matrix Factorization for Face Recognition

  2002cited by this paper
• Algorithms for Non-negative Matrix Factorization

  2000cited by this paper
• Local Uniqueness and Convergence of Iterative Methods for Nonsmooth Variational Inequalities

  1995influential reference
• Minimization of SC1 functions and the Maratos effect

  1995influential reference
• Convergence Analysis of Some Algorithms for Solving Nonsmooth Equations

  1993cited by this paper
• A nonsmooth version of Newton's method

  1993influential reference

• No citing papers are available for this paper.