arXiv · 2611.14345
Smoothed Least-Absolute Conditional Multidimensional Scaling via Iteratively Reweighted Majorization
Abstract
Conditional multidimensional scaling combines latent coordinates with observed covariates to fit pairwise dissimilarities. We develop a smoothed l1 formulation using the shifted Charbonnier loss. We prove residualwise sharpness of the quadratic loss majorizer, monotonicity, compact containment, and stationarity of noncollapsed accumulation points. A uniform additive bound quantifies approximation to exact least-absolute conditional stress. The experiments use paired contamination, separately calibrated loss parameters, combined-configuration recovery, and direct checks of the theorem's inequalities.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Xiao Ling. 2026-11-10. Smoothed Least-Absolute Conditional Multidimensional Scaling via Iteratively Reweighted Majorization. https://arxiv.science/abs/2611.14345
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite arXiv Science by Indiciae: BibTeX · RIS.