Research map / Learning theory and algorithms / Prediction and decision losses
Representation Tradeoffs for Hyperbolic Embeddings
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 the combinatorial tree embedding | Hyperbolic graph embeddings · WordNet | MAP: 0.87* -> 0.989 | Helped |
| Use a learned scale instead of no scale (ablation) | PyTorch embedding · CS PhDs, rank 50 | MAP: 0.481 -> 0.508 | Helped |
| Use a learned scale instead of no scale (ablation) | PyTorch embedding · CS PhDs, rank 100 | MAP: 0.688 -> 0.681 | Hurt |
| Use a learned scale instead of no scale (ablation) | PyTorch embedding · CS PhDs, rank 200 | MAP: 0.894 -> 0.907 | Helped |
| Apply exponential distance weighting (ablation) | PyTorch embedding · CS PhDs, rank 50 | MAP: 0.481 -> 0.775 | Helped |
| Apply exponential distance weighting (ablation) | PyTorch embedding · CS PhDs, rank 100 | MAP: 0.688 -> 0.882 | Helped |
| Apply exponential distance weighting (ablation) | PyTorch embedding · CS PhDs, rank 200 | MAP: 0.894 -> 0.963 | Helped |
| Use the combinatorial construction rather than optimization-based FB | Hyperbolic embedding · Tree-like hierarchies, including WordNet | MAP: 0.87* -> 0.989 | Helped |
Closest papers by meaning
- Tree! I am no Tree! I am a low dimensional Hyperbolic Embedding (NeurIPS 2020 · similarity 0.93)
- Hyperbolic Embeddings of Supervised Models (NeurIPS 2024 · similarity 0.91)
- Low-distortion and GPU-compatible Tree Embeddings in Hyperbolic Space (ICML 2025 · similarity 0.91)
- Weighted Embeddings for Low-Dimensional Graph Representation (AAAI 2025 · similarity 0.90)
- Learning Structured Representations with Hyperbolic Embeddings (NeurIPS 2024 · similarity 0.89)
- Tree-Wasserstein Distance for High Dimensional Data with a Latent Feature Hierarchy (ICLR 2025 · similarity 0.88)
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.