Research map / Learning theory and algorithms / Matrix and tensor methods
A Blind Block Term Decomposition of High Order Tensors
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use algebraic BBTD instead of MINF or NLS | Tensor decomposition on synthetic setting 3 · Synthetic setting 3, η=1e-6 | relative residual: 9.8e-6 -> 9.9e-6 (MINF); 9.8e-6 -> 9.9e-6 (NLS) | No clear effect |
| Use algebraic BBTD instead of MINF or NLS | Tensor decomposition on synthetic setting 3 · Synthetic setting 3, η=1e-3 | CPU time: 0.14 -> 0.07 (MINF); 0.12 -> 0.07 (NLS) | Helped |
| Use algebraic BBTD instead of MINF or NLS | Tensor decomposition on synthetic setting 4 · Synthetic setting 4 | relative residual: 6.4e-2 -> 1.7e-5; 1.0e-1 -> 9.9e-4 (MINF); 8.8e-2 -> 1.7e-5; 1.1e-1 -> 9.9e-4 (NLS) | Helped |
| Use algebraic BBTD instead of MINF or NLS | Tensor decomposition on synthetic setting 4 · Synthetic setting 4 | CPU time: 0.81 -> 0.15; 0.78 -> 0.14 (MINF); 0.99 -> 0.15; 0.89 -> 0.14 (NLS) | Helped |
| Use algebraic BBTD instead of MINF or NLS | Tensor decomposition on synthetic setting 5 · Synthetic setting 5 | relative residual: 2.9e-1 -> 1.1e-6; 3.2e-1 -> 1.1e-3 (MINF); 2.9e-1 -> 1.1e-6; 2.6e-1 -> 1.1e-3 (NLS) | Helped |
| Use algebraic BBTD instead of MINF or NLS | Tensor decomposition on synthetic setting 5 · Synthetic setting 5 | CPU time: 2.40 -> 0.34; 2.39 -> 0.27 (MINF); 26.4 -> 0.34; 26.2 -> 0.27 (NLS) | Helped |
| Increase the number of blocks in the synthetic decomposition | SVDS4BBTD · Synthetic setting 2, R=2, 4, 6, 8, 10 | successful recovery rate: 100 -> 40 | Hurt |
Closest papers by meaning
- Fitting Low-Rank Tensors in Constant Time (NeurIPS 2017 · similarity 0.86)
- Geometric All-way Boolean Tensor Decomposition (NeurIPS 2020 · similarity 0.86)
- Score-Based Model for Low-Rank Tensor Recovery (AAAI 2026 · similarity 0.86)
- Landscape analysis of an improved power method for tensor decomposition (NeurIPS 2021 · similarity 0.85)
- Fast Tensor Completion via Approximate Richardson Iteration (ICML 2025 · similarity 0.85)
- Towards Multi-Mode Outlier Robust Tensor Ring Decomposition (AAAI 2024 · similarity 0.84)
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.