Research map / Learning theory and algorithms / Matrix and tensor methods
Handling The Non-Smooth Challenge in Tensor SVD: A Multi-Objective Tensor Recovery Framework
Jingjing Zheng, Wanglong Lu, Wenzhe Wang, Yankai Cao, Xiaoqin Zhang, Xianta Jiang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace fixed-transform t-SVD (HTNN-DCT) with learnable unitary transforms (TC-SL) on shuffled image sequences | higher-order t-SVD tensor completion · shuffled CIFAR/LFW/GTF (p=0.3) | average PSNR: 24.02 -> 28.20 | Improved |
| add multi-objective learning across dimensions (MOTC vs TC-SL) (ablation) | TC-SL · HMDB51 video inpainting | average PSNR: 36.82 -> 38.74 | Improved |
| add multi-objective learning across dimensions (MOTC vs TC-SL) (ablation) | TC-SL · BSD image inpainting | average PSNR: 32.06 -> 33.58 | Improved |
| use learnable tensor nuclear norm (TC-SL) instead of fixed DCT transform for non-smooth video | HTNN-DCT · HMDB51 video inpainting | average PSNR: 27.71 -> 36.82 | Improved |
| use higher-order tensor methods instead of three-order TNN methods | BSD image inpainting · BSD (p=0.3) | PSNR: 23.25 -> 25.45 (TNN-DCT to KBR) | Improved |
| use MOTC instead of WSTNN weighted-sum across modes | higher-order tensor completion · BSD image inpainting (average) | PSNR: 31.31 -> 33.58 | Improved |
Closest papers by meaning
- Low-Rank Tensor Completion: A Pseudo-Bayesian Learning Approach (ICCV 2017 · similarity 0.91)
- Low-Rank Tensor Completion With a New Tensor Nuclear Norm Induced by Invertible Linear Transforms (CVPR 2019 · similarity 0.90)
- Guaranteed Nonconvex Low-Rank Tensor Estimation via Scaled Gradient Descent (JMLR 2026 · similarity 0.90)
- Guaranteed Tensor Recovery Fused Low-rankness and Smoothness (TPAMI 2023 · similarity 0.90)
- Tensor Ring Decomposition with Rank Minimization on Latent Space: An Efficient Approach for Tensor Completion (AAAI 2019 · similarity 0.90)
- Fast Tensor Completion via Approximate Richardson Iteration (ICML 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
