EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Fast Unconstraint Convex Symmetric Matrix for Semi-Supervised Learning

Wenhao Wang, Kaiwen Chen, Wenjun Luo, Nan Zhou, Yanyi Cao

Symmetry · 21.04.2026 · 18 (4) · 698

DOI: 10.3390/sym18040698 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Symmetry, online erschienen am 21.04.2026. Verfasst von Wenhao Wang und 4 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Sichuan Provincial Department of Science and Technology.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Symmetric matrix factorization (SMF) plays an important role in clustering and representation learning. Nevertheless, most existing SMF-based approaches are formulated as non-convex optimization problems, which often leads to unstable convergence and high computational costs. In this paper, we develop a fast unconstrained convex symmetric matrix factorization framework, termed FUCSMF, for semi-supervised learning. By incorporating label information into the symmetric factorization formulation, the proposed model is transformed into a convex objective, which guarantees global optimality and enables efficient optimization using standard unconstrained solvers. To further improve scalability, a bipartite graph structure is introduced into SMF from a hypergraph-inspired perspective, significantly reducing the computational burden. The resulting computational complexity is reduced to O(nmd), which is substantially lower than the O(nmd+m2n+m3) complexity required by existing bipartite graph-based methods, where n, m, and d denote the numbers of samples, anchor points, and feature dimensions, respectively. In addition, we propose a correntropy-based graph construction strategy to alleviate the sensitivity of conventional adaptive neighbor bipartite graph methods. Extensive experiments on six benchmark datasets, involving comparisons with eleven state-of-the-art methods, demonstrate that FUCSMF achieves superior clustering performance while requiring significantly less computational time. Empirical results further show that the proposed method converges rapidly, typically within ten iterations.

Förderung

  • Sichuan Provincial Department of Science and Technology

Zitieren

APA

Wang, W., Chen, K., Luo, W., Zhou, N., & Cao, Y. (2026). Fast Unconstraint Convex Symmetric Matrix for Semi-Supervised Learning. Symmetry, 18(4), 698. https://doi.org/10.3390/sym18040698

Vancouver

Wang W, Chen K, Luo W, Zhou N, Cao Y. Fast Unconstraint Convex Symmetric Matrix for Semi-Supervised Learning. Symmetry. 2026;18(4):698. doi:10.3390/sym18040698

BibTeX
@article{Wang2026,
  author = {Wang, Wenhao and Chen, Kaiwen and Luo, Wenjun and Zhou, Nan and Cao, Yanyi},
  title = {Fast Unconstraint Convex Symmetric Matrix for Semi-Supervised Learning},
  journal = {Symmetry},
  year = {2026},
  volume = {18},
  number = {4},
  pages = {698},
  doi = {10.3390/sym18040698}
}
Kostenlos starten →