Research map / Learning theory and algorithms / Prediction and decision losses
Towards Optimal Binary Code Learning via Ordinal Embedding
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 OEH ordinal-embedding hashing instead of ITQ | unsupervised hashing · CIFAR10 8 bits | mAP: 0.1512 -> 0.1682 | Helped |
| use OEH instead of ITQ at 8 bits | unsupervised hashing · LabelMe 8 bits | mAP: 0.3516 -> 0.3022 | Hurt |
| use OEH instead of ITQ at 64 bits | unsupervised hashing · LabelMe 64 bits | mAP: 0.4467 -> 0.4512 | Helped |
| use OEH instead of DTW-only matching in query by humming | query by humming · MIR-QBSH | query time: 17.071 -> 0.695 s (OEH+DTW) | Helped |
| replace EMD with OEH as first-stage filter before DTW | query by humming · MIR-QBSH | Top-10 hit rate: 0.914 -> 0.907 | No clear effect |
| replace LSH/ITQ with OEH as first-stage filter | OEH+DTW · MIR-QBSH | Top-10 hit rate: 0.586 (LSH) / 0.872 (ITQ) -> 0.907 | Helped |
| increase number of nearest neighbours k from 10 to 20 (ablation) | OEH · CIFAR10 and LabelMe 16 bits | mAP | Helped |
Closest papers by meaning
- Ordinal Constrained Binary Code Learning for Nearest Neighbor Search (AAAI 2017 · similarity 0.92)
- Efficient end-to-end learning for quantizable representations (ICML 2018 · similarity 0.87)
- Discretely Coding Semantic Rank Orders for Supervised Image Hashing (CVPR 2017 · similarity 0.86)
- Hashing as Tie-Aware Learning to Rank (CVPR 2018 · similarity 0.86)
- Statistical Model-driven Similarity Hashing: Bridging Modalities for Efficient Unsupervised Retrieval (AAAI 2025 · similarity 0.84)
- Pairwise-Label-Based Deep Incremental Hashing with Simultaneous Code Expansion (AAAI 2024 · similarity 0.84)
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.