A Joint Sparse Self-Representation Learning Method for Multiview Clustering

2025-08-09

AI Summary

To address the challenges of imprecise local structure extraction and inconsistent global affinity matrix construction in multi-view clustering, this paper proposes a joint sparse self-representation learning framework. Specifically, it replaces conventional graph Laplacian regularization with an ℓ₀-norm constraint to precisely capture view-specific local structures, and incorporates a low-rank constraint to jointly learn a consensus affinity matrix that encodes cross-view global consistency. To solve the resulting nonconvex and nonsmooth optimization problem, we design an alternating quadratic penalty algorithm with guaranteed global convergence, where all subproblems admit closed-form solutions. Extensive experiments on six benchmark datasets demonstrate that the proposed method significantly outperforms eight state-of-the-art algorithms, achieving both superior clustering accuracy and enhanced optimization stability.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Non-convex Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Multiview clustering (MC) aims to group samples using consistent and complementary information across various views. The subspace clustering, as a fundamental technique of MC, has attracted significant attention. In this paper, we propose a novel joint sparse self-representation learning model for MC, where a featured difference is the extraction of view-specific local information by introducing cardinality (i.e., $ell_0$-norm) constraints instead of Graph-Laplacian regularization. Specifically, under each view, cardinality constraints directly restrict the samples used in the self-representation stage to extract reliable local and global structure information, while the low-rank constraint aids in revealing a global coherent structure in the consensus affinity matrix during merging. The attendant challenge is that Augmented Lagrange Method (ALM)-based alternating minimization algorithms cannot guarantee convergence when applied directly to our nonconvex, nonsmooth model, thus resulting in poor generalization ability. To address it, we develop an alternating quadratic penalty (AQP) method with global convergence, where two subproblems are iteratively solved by closed-form solutions. Empirical results on six standard datasets demonstrate the superiority of our model and AQP method, compared to eight state-of-the-art algorithms.

Problem

• Improving multiview clustering via joint sparse self-representation learning
• Extracting reliable local and global structure information without Graph-Laplacian
• Ensuring convergence in nonconvex models with alternating quadratic penalty method

Innovation

• Joint sparse self-representation learning for clustering
• Cardinality constraints replace Graph-Laplacian regularization
• Alternating quadratic penalty method ensures global convergence