Research map / Learning theory and algorithms / Kernels and regression theory
Kernelized Support Tensor Machines
Lifang He, Chun-Ta Lu, Guixiang Ma, Shen Wang, Linlin Shen, Philip S. Yu, Ann B. Ragin
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 |
|---|---|---|---|
| KSTM vs best baseline MMKbest | KSTM · ADNI | accuracy: 0.81 ± 0.01 -> 0.84 ± 0.03 | Improved |
| KSTM vs best baseline MMKbest | KSTM · HIV | accuracy: 0.79 ± 0.01 -> 0.82 ± 0.02 | Improved |
| KSTM vs best baseline MMKbest | KSTM · ADHD | accuracy: 0.70 ± 0.01 -> 0.74 ± 0.02 | Improved |
| supervised label-aware factorization (KSTM) vs unsupervised CP-kernel DuSK | KSTM · ADNI | accuracy: 0.75 ± 0.02 -> 0.84 ± 0.03 | Improved |
| kernel factorization with label information (KSTM) vs MMK | KSTM · ADNI | accuracy: 0.81 ± 0.01 -> 0.84 ± 0.03 | Improved |
| preserve tensor structure (DuSK) instead of unfolding into vectors (SVM) | kernel classifier · ADNI | accuracy: 0.49 ± 0.02 -> 0.75 ± 0.02 | Improved |
| PCA before SVM | SVM · ADHD | accuracy: 0.58 ± 0.00 -> 0.63 ± 0.01 | Improved |
| MMK with best random shared kernel (50 tries) instead of covariance-based | MMK · ADNI | accuracy: 0.69 ± 0.01 -> 0.81 ± 0.01 | Improved |
| 3D CNN vs KSTM | 3D CNN · ADNI | accuracy: 0.84 ± 0.03 -> 0.52 ± 0.03 | Hurt |
| vary factorization rank R from 1 to 10 (ablation) | KSTM · ADNI, HIV, ADHD | test accuracy | Mixed |
| optimal R range (ablation) | KSTM · three datasets | test accuracy | Improved |
| K3rd tensor kernel vs SVM | K3rd · HIV | accuracy: 0.70 ± 0.01 -> 0.75 ± 0.02 | Improved |
Closest papers by meaning
- Efficient Structure-preserving Support Tensor Train Machine (JMLR 2023 · similarity 0.90)
- Multi-Way Multi-Level Kernel Modeling for Neuroimaging Classification (CVPR 2017 · similarity 0.90)
- Factor Augmented Tensor-on-Tensor Neural Networks (AAAI 2025 · similarity 0.87)
- Large-scale Multi-view Tensor Clustering with Implicit Linear Kernels (CVPR 2025 · similarity 0.85)
- High-Dimensional Tensor Regression With Oracle Properties (ICML 2025 · similarity 0.84)
- Over-parameterized Student Model via Tensor Decomposition Boosted Knowledge Distillation (NeurIPS 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.
