
Mathematical Proofs for SPD Inner Products and Pseudo-Gyrodistances in Manifold Layers
4 Dec 2024
Detailed proofs for SPD spaces, inner products, pseudo-gyrodistances, and FC layers, referencing Nguyen & Yang (2023) and Pennec et al. (2020).

Canonical Representations and Computation in SPD Matrices
4 Dec 2024
Learn to Compute canonical representations for SPD matrices using orthonormal matrices & SVD, enabling efficient geometric applications and matrix manipulation

What Are Gyrocommutative Gyrogroups?
3 Dec 2024
Gyrogroups, gyrocommutative gyrogroups, and gyrovector spaces are key to hyperbolic geometry and AI, offering insights into advanced mathematical structures

The Limitations of GyroSpd++ and Gr-GCN++ in Human Action Recognition and Graph Embedding Tasks
3 Dec 2024
GyroSpd++ and Gr-GCN++ face limitations with hybrid methods and different Riemannian metrics, suggesting a need for optimization in SPD and Grassmann networks.

Evaluating Gr-GCN++ for Node Classification Across Various Datasets: Results and Comparisons
3 Dec 2024
Gr-GCN++ outperforms competitors in node classification across Airport, Pubmed, and Cora datasets, highlighting the value of perspective in Grassmann manifolds

Exploring the Impact of Riemannian Metrics in Human Action Recognition Tasks Using GyroSpd++
3 Dec 2024
Evaluate GyroSpd++ for human action recognition using HDM05, FPHA, NTU60 datasets, with insights on performance, design, and comparisons to other methods

Key Notations and Algorithm for Computing Pseudo-Gyrodistances in Structure Spaces
2 Dec 2024
Learn notations and algorithms for computing pseudo-gyrodistances, crucial for MLR in Riemannian manifold-based neural networks.

Bridging Geometry and Deep Learning: Key Developments in SPD and Grassmann Networks
2 Dec 2024
This paper presents novel advancements in SPD neural networks and Grassmann geometry for action recognition and node classification.

New Riemannian Networks Outperform Traditional Models in Action Recognition and Node Classification
2 Dec 2024
GyroSpd++ and Gr-GCN++ outperform baselines in human action recognition and node classification, showing superior accuracy on NTU60, FPHA, and Pubmed datasets.