AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Near Neighbor: Who is the Fairest of Them All?

Sariel Har-Peled, Sepideh Mahabadi

NeurIPS 2019 · Learning theory and algorithms › Prediction and decision losses · approach: Approximate nearest neighbor indexing · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (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 changedWhereMeasuredEffect
Use degree approximation instead of the standard bucket-sampling baselinesFair near-neighbor sampling · MNISTTV distance ratio vs. Optimal: 2.4 vs. 6.6 and 10Improved
Use degree approximation instead of the standard bucket-sampling baselinesFair near-neighbor sampling · SIFT10KTV distance ratio vs. Optimal: 1.4 vs. 6.1 and 9.7Improved
Use degree approximation instead of the standard bucket-sampling baselinesFair near-neighbor sampling · GloVeTV distance ratio vs. Optimal: 2.7 vs. 6.5 and 13.1Improved
Use degree approximation rather than exact degree computationFair near-neighbor sampling · MNIST, k=15, L=100runtime: twice fasterImproved
Use degree approximation rather than exact degree computationFair near-neighbor sampling · MNIST, L=300runtime: 4.3 times fasterImproved
Spend twice more queries on degree approximation (ablation)Degree approximation · MNISTTV distance ratio vs. Optimal: 2.4 to 1.6Improved
Spend three times more queries on degree approximation (ablation)Degree approximation · MNISTTV distance ratio vs. Optimal: to 1.2Improved
Spend four times more queries on degree approximation (ablation)Degree approximation · MNISTTV distance ratio vs. Optimal: to 1.05Improved
Spend twice more queries on degree approximation (ablation)Degree approximation · SIFT10KTV distance ratio vs. Optimal: 1.4 to 1.16Improved
Spend three times more queries on degree approximation (ablation)Degree approximation · SIFT10KTV distance ratio vs. Optimal: to 1.04Improved
Spend four times more queries on degree approximation (ablation)Degree approximation · SIFT10KTV distance ratio vs. Optimal: to 1.05Improved
Spend twice more queries on degree approximation (ablation)Degree approximation · GloVeTV distance ratio vs. Optimal: 2.7 to 1.47Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy