Robust Subspace-Constrained Quadratic Models for Low-Dimensional Structure Learning
The paper presents a robust subspace-constrained quadratic model for learning low-dimensional structures from high-dimensional data. It enhances the existing framework to handle various noise distributions, improving robustness and reconstruction accuracy. The authors also introduce a gradient-based algorithm for efficient optimization and provide a sensitivity analysis of different loss functions.
- ▪The proposed model accommodates a broad class of noise distributions, including generalized Gaussian and radial Laplace models.
- ▪A gradient-based algorithm with a backtracking line-search strategy is developed for stable and efficient convergence.
- ▪Extensive numerical experiments show that the proposed approach outperforms existing methods in terms of robustness and reconstruction accuracy.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Machine Learning arXiv:2605.20300 (cs) [Submitted on 19 May 2026] Title:Robust Subspace-Constrained Quadratic Models for Low-Dimensional Structure Learning Authors:Zheng Zhai, Xiaohui Li View a PDF of the paper titled Robust Subspace-Constrained Quadratic Models for Low-Dimensional Structure Learning, by Zheng Zhai and Xiaohui Li View PDF HTML (experimental) Abstract:In this paper, we propose a robust subspace-constrained quadratic model (SCQM) for learning low-dimensional structure from high-dimensional data. Building upon the subspace-constrained quadratic matrix factorization (SQMF) framework, the proposed model accommodates a broad class of noise distributions, including generalized Gaussian and radial Laplace models.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
