Research map / Learning theory and algorithms / Prediction and decision losses
Boosting Complementary Hash Tables for Fast Nearest Neighbor Search
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 |
|---|---|---|---|
| boosted joint multi-table learning vs sequential heuristic CH | multi-table hashing · SIFT-1M L=16 | average precision (%): 40.03 -> 46.17 | Helped |
| BCH vs CH with 4 tables | multi-table hashing · SIFT-1M L=4 | average precision (%): 38.45 -> 41.30 | Helped |
| BCH vs CH | multi-table hashing · GIST-1M L=16 | average precision (%): 16.03 -> 17.15 | Helped |
| BCH vs CH | multi-table hashing · NUS-WIDE L=16 | average precision (%): 37.43 -> 39.10 | Helped |
| BCH vs ITQ | multi-table hashing · CIFAR-10 L=16 | average precision (%): 21.26 -> 26.13 | Helped |
| increase number of tables from 4 to 16 | BCH · SIFT-1M | average precision (%): 41.30 -> 46.17 | Helped |
| increase number of tables from 4 to 16 | MVCH · SIFT-1M | average precision (%): 17.75 -> 15.73 | Hurt |
| BCH vs MVCH training cost | multi-table hashing · SIFT-1M | training time (s): 1106.85 -> 153.52 | Helped |
| BCH vs CH training cost | multi-table hashing · SIFT-1M | training time (s): 53.20 -> 153.52 | Hurt |
| BCH vs CH on GIST-1M with a single hash table | BCH · SIFT-1M | F1-measure | Hurt |
Closest papers by meaning
- Fast Online Hashing with Multi-Label Projection (AAAI 2023 · similarity 0.87)
- Multilinear Hyperplane Hashing (CVPR 2016 · similarity 0.86)
- An ensemble diversity approach to supervised binary hashing (NeurIPS 2016 · similarity 0.86)
- Learning to Hash Robustly, Guaranteed (ICML 2022 · similarity 0.85)
- Supervised Hashing via Uncorrelated Component Analysis (AAAI 2016 · similarity 0.85)
- KALAHash: Knowledge-Anchored Low-Resource Adaptation for Deep Hashing (AAAI 2025 · 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.