AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

A Universal Nearest-Neighbor Estimator for Intrinsic Dimensionality

Eng-Jon Ong, Omer Bobrowski, Gesine Reinert, Primož Škraba

TPAMI 2026 · Learning theory and algorithms › Sample complexity · approach: Density and property estimation · 3 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
add parameter optimization on the benchmark (alpha, beta tuned) to L2N2(2,1) (ablation)L2N2(2,1) · Benchmark manifolds (2,500 pts)MPE (%): 9.96 -> 8.09Improved
round estimated dimensions to nearest integer (ablation)L2N2(2,1) · Benchmark manifolds (2,500 pts)MPE (%): 9.96 -> 6.06Improved
use larger neighbor indices (k,j)=(8,4) instead of (2,1) (ablation)L2N2 · Benchmark manifolds (2,500 pts)MPE (%): 9.96 -> 12.96Hurt
use (4,2) instead of (2,1) (ablation)L2N2 · Benchmark manifolds (2,500 pts)MPE (%): 9.96 -> 11.38Hurt
use L2N2(2,1) instead of TwoNNTwoNN · Benchmark manifolds (2,500 pts)MPE (%): 10.99 -> 9.96Improved
use L2N2(2,1) instead of MLEMLE · Benchmark manifolds (2,500 pts)MPE (%): 15.96 -> 9.96Improved
increase sample size from 625 to 5,000 pointsL2N2(2,1) · Benchmark manifoldsMPE (%): 13.02 -> 9.28Improved
subsample query points to 2,500 for large datasets (ablation)L2N2(2,1) · Benchmark manifolds (500k pts)MPE (%): 6.41 -> 6.88Hurt
add Gaussian ambient noise sigma 0.1 to S^6L2N2(2,1) · S^6 in R^11estimated dimension: 6.10 -> 8.48Hurt
use L2N2 estimate as autoencoder bottleneck instead of TwoNNautoencoder · MNISTreconstruction MSE: 0.0143 -> 0.0119Improved
use dimension range 10-40 instead of 1-20 when fitting alpha at n=2,500 (ablation)L2N2 tuning · Gaussian calibrationalpha_{2,1}: 0.929932 -> 0.762282Hurt

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