Research map / Learning theory and algorithms / Matrix and tensor methods
The Gyro-Structure of Some Matrix Manifolds
Xuan Son Nguyen
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 |
|---|---|---|---|
| use GyroAI-HAUNet instead of SPDNetBN at M = 3 | SPDNetBN · FPHA | accuracy: 91.83 -> 96.00 | Improved |
| use GyroAI-HAUNet instead of SPDNetBN at M = 3 | SPDNetBN · NTU60 | accuracy: 79.52 -> 94.72 | Improved |
| use GyroAI-HAUNet instead of SPDNetBN at M = 3 | SPDNetBN · HDM05 | accuracy: 76.25 -> 78.14 | Improved |
| use GyroAI-HAUNet instead of SPDNetBN at M = 1 | SPDNetBN · HDM05 | accuracy: 62.54 -> 61.50 | Hurt |
| use GyroAI-HAUNet instead of SPDNet at M = 1 | SPDNet · HDM05 | accuracy: 58.44 -> 61.50 | Improved |
| use GyroLE-HAUNet instead of GyroAI-HAUNet | GyroAI-HAUNet · NTU60 M = 3 | accuracy: 94.72 -> 89.44 | Hurt |
| use GyroLE-HAUNet instead of GyroAI-HAUNet | GyroAI-HAUNet · FPHA M = 3 | accuracy: 96.00 -> 89.94 | Hurt |
| increase temporal pyramid M from 1 to 3 | GyroAI-HAUNet · FPHA | accuracy: 89.73 -> 96.00 | Improved |
| increase temporal pyramid M from 1 to 3 | GyroAI-HAUNet · HDM05 | accuracy: 61.50 -> 78.14 | Improved |
| use Grassmann GyroGR-QANet instead of SPD SPDR Sca | SPDR Sca · TrecQA | MAP: 47.77 ± 0.18 -> 50.18 ± 1.29 | Improved |
| use Grassmann GyroGR-QANet instead of SPD SPDR Sca | SPDR Sca · WikiQA | MAP: 59.42 ± 0.04 -> 56.69 ± 1.45 | Hurt |
| product manifold GyroGR-SPDF1 Sca instead of Grassmann only | GyroGR-QANet · WikiQA | MAP: 56.69 ± 1.45 -> 60.62 ± 0.25 | Improved |
Closest papers by meaning
- Matrix Manifold Neural Networks++ (ICLR 2024 · similarity 0.92)
- Neural networks on Symmetric Spaces of Noncompact Type (ICLR 2025 · similarity 0.86)
- Building Transformation Layers for Riemannian Neural Networks (NeurIPS 2026 · similarity 0.85)
- Understanding Matrix Function Normalizations in Covariance Pooling through the Lens of Riemannian Geometry (ICLR 2025 · similarity 0.85)
- Siegel Neural Networks (NeurIPS 2025 · similarity 0.84)
- Spiking Graph Neural Network on Riemannian Manifolds (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.
