Research map / Label-efficient and robust learning / Recognition and long-tail
Learning Multi-Way Relations via Tensor Decomposition With Neural Networks
Koji Maruhashi, Masaru Todoriki, Takuya Ohwa, Keisuke Goto, Yu Hasegawa, Hiroya Inakoshi, Hirokazu Anai
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 Tucker decomposition with SRTD (ablation) | DeepTensor · IDS probing | average precision: 62.10±5.05 -> 78.52±3.65 | Improved |
| replace Tucker decomposition with SRTD (ablation) | DeepTensor · IDS DOS | average precision: 48.75±2.59 -> 62.18±3.56 | Improved |
| replace Tucker decomposition with SRTD (ablation) | DeepTensor · LEND | average precision: 4.99±1.15 -> 12.07±1.40 | Improved |
| replace Tucker decomposition with SRTD (ablation) | DeepTensor · ENZYMES | accuracy: 21.93±0.44 -> 47.17±1.46 | Improved |
| learn target core tensor with EBP (vs fixed random) (ablation) | DeepTensor · IDS probing | average precision: 66.16±10.84 -> 78.52±3.65 | Improved |
| learn target core tensor with EBP (vs fixed random) (ablation) | DeepTensor · IDS DOS | average precision: 61.00±5.27 -> 62.18±3.56 | Improved |
| learn target core tensor with EBP (vs fixed random) (ablation) | DeepTensor · LEND | average precision: 4.78±3.42 -> 12.07±1.40 | Improved |
| learn target core tensor with EBP (vs fixed random) (ablation) | DeepTensor · ENZYMES | accuracy: 47.07±0.76 -> 47.17±1.46 | No clear effect |
| learn target core tensor with EBP (vs fixed random) (ablation) | DeepTensor · NCI1 | accuracy: 75.15±0.41 -> 75.22±0.30 | No clear effect |
| add hidden layers to the classifier (ablation) | DeepTensor · IDS probing | average precision: 36.03±5.72 -> 78.52±3.65 | Improved |
| add hidden layers to the classifier (ablation) | DeepTensor · LEND | average precision: 10.10±8.51 -> 12.07±1.40 | Mixed |
| add hidden layers to the classifier (ablation) | DeepTensor · NCI1 | accuracy: 70.42±0.62 -> 75.22±0.30 | Improved |
Closest papers by meaning
- Learning from Multiway Data: Simple and Efficient Tensor Regression (ICML 2016 · similarity 0.86)
- Multi-Way Multi-Level Kernel Modeling for Neuroimaging Classification (CVPR 2017 · similarity 0.84)
- Efficient Structure-preserving Support Tensor Train Machine (JMLR 2023 · similarity 0.83)
- On the Joint Interaction of Models, Data, and Features (ICLR 2024 · similarity 0.82)
- Generalized Tensor Decomposition for Understanding Multi-Output Regression under Combinatorial Shifts (NeurIPS 2024 · similarity 0.81)
- Large-scale Multi-view Tensor Clustering with Implicit Linear Kernels (CVPR 2025 · similarity 0.81)
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.
