Robust Low-Tubal-Rank Tensor Completion under Cross-Concentrated Sampling

DOI10.48550/arXiv.2608.03928MaRDI QIDQ7284591arXivFDO

Authors HanQin Cai, Longxiu Huang, Jing Qin, Chengyue Wu

Publication date 4 August 2026

Published in arXiv

Tldr We propose R-ItCUR, a tensor-native algorithm for robust low-tubal-rank tensor completion from partially observed tensors contaminated by sparse outliers via t-CCS. We apply adaptive blockwise Welsch correction and projected blockwise gradient descent to update the low-rank component directly on the sampled cross, avoiding full tensor reconstruction and achieving substantial memory and computational savings.

Full work available at URL https://doi.org/10.48550/arXiv.2608.03928

arXiv classification

cs.LGstat.MLmath.NAcs.IT

Cites work

• Accelerated Alternating Projections for Robust Principal Component Analysis
• Accelerating Ill-Conditioned Low-Rank Matrix Estimation via Scaled Gradient Descent
• Accelerating ill-conditioned Hankel matrix recovery via structured Newton-like descent
• Adaptive Cross Tubal Tensor Approximation
• CUR matrix decompositions for improved data analysis
• Correction of Corrupted Columns Through Fast Robust Hankel Matrix Completion
• Exact Tensor Completion Using t-SVD
• Factorization strategies for third-order tensors
• Fast Algorithms for Robust PCA via Gradient Descent
• Learned Robust PCA: A Scalable Deep Unfolding Approach for High-Dimensional Outlier Detection
• Low-rank Matrix Recovery from Errors and Erasures
• Matrix Completion with Cross-Concentrated Sampling: Bridging Uniform Sampling and CUR Sampling
• Mode-wise Tensor Decompositions: Multi-dimensional Generalizations of CUR Decompositions
• Non-convex Robust PCA
• Nonconvex Robust Low-rank Matrix Recovery
• Optimal matrix-mimetic tensor algebras via variable projection
• Perspectives on CUR Decompositions
• Relative-Error CUR Matrix Decompositions
• Robust Matrix Completion via Maximum Correntropy Criterion and Half Quadratic Optimization
• Robust Principal Component Analysis?
• Robust Spectral Compressed Sensing via Structured Matrix Completion
• Robust Tensor CUR Decompositions: Rapid Low-Tucker-Rank Tensor Recovery with Sparse Corruption
• Robust low-rank tensor recovery: models and algorithms
• Robust matrix completion
• Robust principal component analysis?
• Robust regression using iteratively reweighted least-squares
• Robust tensor CUR decompositions: rapid low-Tucker-rank tensor recovery with sparse corruptions
• Robust tensor completion: equivalent surrogates, error bounds, and algorithms
• Stability of Sampling for CUR Decompositions
• Structured Gradient Descent for Fast Robust Low-Rank Hankel Matrix Completion
• Tensor CUR decomposition under T-product and its perturbation
• Variable density sampling with continuous trajectories

This page was built for publication: Robust Low-Tubal-Rank Tensor Completion under Cross-Concentrated Sampling

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7284591)