Research map / Learning theory and algorithms / Prediction and decision losses
Scaling Up Ordinal Embedding: A Landmark Approach
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 L-SOE triplets (LNM-MDS landmark triplets) instead of random triplets for SOE embedding (ablation) | SOE / L-SOE medium-scale embedding · Synthetic Uniform Ball and GMM, n up to 16,000, R30 | convergence speed and embedding quality | Mixed |
| Embed remaining points independently via landmark shell constraints (LLOE phase) rather than a joint optimization | LLOE large-scale embedding · Synthetic Uniform Ball and GMM, n up to 1.2M, R30, m=1000 | median CPU time and distance error | Helped |
| Use LLOE landmark-based embedding instead of joint SOE optimization at large n | large-scale ordinal embedding · Synthetic data, d=5 | points embedded per second: ~1,100 vs 16 (L-SOE) and 670 (random SOE) points/s | Helped |
| Replace LSA-like exact-distance baseline comparison: LLOE vs LSA on 20 Newsgroups and MNIST | LLOE · 20 Newsgroups and MNIST Digits, R30 | order error probability and NN error | Mixed |
| Use farthest-first traversal ordering for anchor selection in LLOE | LLOE · Synthetic data | anchor selection quality | No clear effect |
| Add random triplets to L-SOE triplet set (ablation) | L-SOE · Synthetic data | embedding time | No clear effect |
Closest papers by meaning
- On the rankability of visual embeddings (NeurIPS 2025 · similarity 0.88)
- Euclidean Embedding of Data Using Local Distances (NeurIPS 2026 · similarity 0.87)
- Embedding Dimension of Contrastive Learning and $k$-Nearest Neighbors (NeurIPS 2024 · similarity 0.86)
- A general kernel framework for non-CND distance measures using $|\mathcal{D}|$-dimensional sparse landmark embeddings (NeurIPS 2026 · similarity 0.85)
- Metric Space Magnitude for Evaluating the Diversity of Latent Representations (NeurIPS 2024 · similarity 0.85)
- Fully Dynamic Embedding into $\ell_p$ Spaces (ICML 2025 · similarity 0.85)
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.