Two rounds of linear-sum assignment identify near neighbors, a three-equation system estimates per-point noise, and subtracting the estimates from corrupted distances yields provably consistent Euclidean distances.
Robust inference of manifold density and geometry by doubly stochas- tic scaling.SIAM Journal on Mathematics of Data Science, 5(3):589–614, 2023
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
stat.ML 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Euclidean Distance Deflation Under High-Dimensional Heteroskedastic Noise
Two rounds of linear-sum assignment identify near neighbors, a three-equation system estimates per-point noise, and subtracting the estimates from corrupted distances yields provably consistent Euclidean distances.