Research map / Learning theory and algorithms / Optimal transport
Scalable Nearest Neighbor Search for Optimal Transport
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 Flowtree instead of Quadtree's tree-distance flow cost, evaluating the tree-optimal flow in the original ground metric | W1 nearest-neighbor search · Real-world datasets | recall@m accuracy | Helped |
| Use Flowtree rather than the quadratic-time baselines | W1 nearest-neighbor search · Real-world datasets | runtime: 5.5 to 30 times faster | Helped |
| Add Flowtree to the tuned pipeline | Quadtree-based 20news nearest-neighbor pipeline · 20news recall@1 >=0.9 | runtime: 0.221s -> 0.059s | Helped |
| Add Flowtree to the tuned pipeline | Quadtree-based 20news nearest-neighbor pipeline · 20news recall@5 >=0.9 | runtime: 0.200s -> 0.027s | Helped |
| Use Flowtree instead of Quadtree | W1 nearest-neighbor search · Random model | nearest-neighbor success probability as N grows | Helped |
| Use Flowtree rather than Quadtree for approximate nearest-neighbor selection | W1 nearest-neighbor search · Uniform distributions | approximation factor: O(log(sn)log(d\Phi)) -> O(log^2 s) | Helped |
| Increase ACT iterations beyond 1 (ablation) | ACT · 20news, Amazon, and MNIST | accuracy | No clear effect |
| Increase Sinkhorn iterations beyond the evaluated 1 or 3 settings (ablation) | Sinkhorn · 20news, Amazon, and MNIST | accuracy | No clear effect |
Closest papers by meaning
- Graph-Based Algorithms for Diverse Similarity Search (ICML 2025 · similarity 0.86)
- UltraTWD: Optimizing Ultrametric Trees for Tree-Wasserstein Distance (ICML 2025 · similarity 0.86)
- Tree-Sliced Wasserstein Distance: A Geometric Perspective (ICML 2025 · similarity 0.85)
- Wasserstein Wormhole: Scalable Optimal Transport Distance with Transformer (ICML 2024 · similarity 0.85)
- Scalable Approximation Algorithms for $p$-Wasserstein Distance and Its Variants (ICML 2025 · similarity 0.85)
- Learning Ultrametric Trees for Optimal Transport Regression (AAAI 2024 · 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.