Research map / Learning theory and algorithms / Matrix and tensor methods
A Statistical Recurrent Model on the Manifold of Symmetric Positive Definite Matrices
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 SPD-SRU instead of TT-GRU | Moving-pattern classification · Moving MNIST, orientation 10-15 | 10-fold testing accuracy: 0.52 ± 0.04 -> 0.96 ± 0.02 | Helped |
| Use SPD-SRU instead of TT-GRU | Moving-pattern classification · Moving MNIST, orientation 10-15-20 | 10-fold testing accuracy: 0.47 ± 0.03 -> 0.94 ± 0.02 | Helped |
| Use SPD-SRU instead of SRU | Moving-pattern classification · Moving MNIST, orientation 10-15 | 10-fold testing accuracy: 0.75 ± 0.19 -> 0.96 ± 0.02 | Helped |
| Use SPD-SRU instead of TT-GRU | Action classification · UCF11 | testing accuracy: 0.78 -> 0.78 | No clear effect |
| Use SPD-SRU instead of TT-GRU | Action classification · UCF11 | number of parameters: 6048 -> 3337 | Helped |
| Use three-layer SPD-SRU models instead of tract Fréchet means with Cramer's test | PD versus control tract testing · LM1 tract | permutation p-value: 0.17 -> 0.01 | Mixed |
| Use three-layer SPD-SRU models instead of tract Fréchet means with Cramer's test | PD versus control tract testing · RM1 tract | permutation p-value: 0.34 -> 0.032 | Mixed |
Closest papers by meaning
- Consistent Geometric Deep Learning via Hilbert Bundles and Cellular Sheaves (NeurIPS 2026 · similarity 0.88)
- Matrix Manifold Neural Networks++ (ICLR 2024 · similarity 0.88)
- Building Transformation Layers for Riemannian Neural Networks (NeurIPS 2026 · similarity 0.88)
- Residual Deep Gaussian Processes on Manifolds (ICLR 2025 · similarity 0.88)
- A Manifold Perspective on the Statistical Generalization of Graph Neural Networks (ICML 2025 · similarity 0.88)
- Geometry-Aware Similarity Metrics for Neural Representations on Riemannian and Statistical Manifolds (NeurIPS 2026 · similarity 0.87)
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.