Research map / Label-efficient and robust learning / Recognition and long-tail
Learning Hyperbolic Representations of Topological Features
Panagiotis Kyriakis, Iordanis Fostiropoulos, Paul Bogdan
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 the Poincare ball instead of Euclidean space (ablation) | P-Poinc · IMDB-M | accuracy: $47.87_{\pm 2.03}$ -> $57.31_{\pm 4.27}$ | Improved |
| Add the learnable parameterization rather than simple parameter addition (ablation) | P-Eucl · IMDB-M | accuracy: $46.45_{\pm 4.03}$ -> $47.87_{\pm 2.03}$ | Improved |
| Use the Poincare ball instead of Euclidean space (ablation) | P-Poinc · IMDB-B | accuracy: $72.48_{\pm 4.57}$ -> $81.86_{\pm 4.26}$ | Improved |
| Add the learnable parameterization rather than simple parameter addition (ablation) | P-Eucl · IMDB-B | accuracy: $67.54_{\pm 3.54}$ -> $72.48_{\pm 4.57}$ | Improved |
| Use the Poincare ball instead of Euclidean space (ablation) | P-Poinc · REDDIT-B | accuracy: $72.87_{\pm 1.75}$ -> $79.78_{\pm 3.21}$ | Improved |
| Add the learnable parameterization rather than simple parameter addition (ablation) | P-Eucl · REDDIT-B | accuracy: $71.45_{\pm 2.98}$ -> $72.87_{\pm 1.75}$ | Improved |
| Use the Poincare ball instead of Euclidean space (ablation) | P-Poinc · REDDIT-5K | accuracy: $46.85_{\pm 2.17}$ -> $51.71_{\pm 3.01}$ | Improved |
| Add the learnable parameterization rather than simple parameter addition (ablation) | P-Eucl · REDDIT-5K | accuracy: $43.15_{\pm 3.12}$ -> $46.85_{\pm 2.17}$ | Improved |
| Use the Poincare ball instead of Euclidean space (ablation) | P-Poinc · REDDIT-12K | accuracy: $36.42_{\pm 4.08}$ -> $42.16_{\pm 3.45}$ | Improved |
| Add the learnable parameterization rather than simple parameter addition (ablation) | P-Eucl · REDDIT-12K | accuracy: $32.56_{\pm 3.68}$ -> $36.42_{\pm 4.08}$ | Improved |
| Include essential features as a separate birth-time histogram | Essential Features Separation baseline · REDDIT-5K | accuracy: $39.30_{\pm 1.12}$ -> $38.45_{\pm 0.98}$ | Hurt |
| Include essential features as a separate birth-time histogram | Essential Features Separation baseline · IMDB-M | accuracy: $38.43_{\pm 0.98}$ -> $38.78_{\pm 0.85}$ | No clear effect |
Closest papers by meaning
- Polynomial Representation for Persistence Diagram (CVPR 2019 · similarity 0.90)
- Perturbation Robust Representations of Topological Persistence Diagrams (ECCV 2018 · similarity 0.90)
- Wasserstein convergence of Cech persistence diagrams for samplings of submanifolds (NeurIPS 2024 · similarity 0.87)
- Point-Level Topological Representation Learning on Point Clouds (ICML 2025 · similarity 0.87)
- Cover learning for large-scale topology representation (ICML 2025 · similarity 0.87)
- Towards Scalable Topological Regularizers (ICLR 2025 · 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.
