pith. sign in

arxiv: 1811.00115 · v1 · pith:EQYIWNXZnew · submitted 2018-10-31 · 📊 stat.ML · cs.LG

Dimensionality Reduction has Quantifiable Imperfections: Two Geometric Bounds

classification 📊 stat.ML cs.LG
keywords precisionmapsmeasurecontinuousdimensionalitydistanceinformationoptimization
0
0 comments X
read the original abstract

In this paper, we investigate Dimensionality reduction (DR) maps in an information retrieval setting from a quantitative topology point of view. In particular, we show that no DR maps can achieve perfect precision and perfect recall simultaneously. Thus a continuous DR map must have imperfect precision. We further prove an upper bound on the precision of Lipschitz continuous DR maps. While precision is a natural measure in an information retrieval setting, it does not measure `how' wrong the retrieved data is. We therefore propose a new measure based on Wasserstein distance that comes with similar theoretical guarantee. A key technical step in our proofs is a particular optimization problem of the $L_2$-Wasserstein distance over a constrained set of distributions. We provide a complete solution to this optimization problem, which can be of independent interest on the technical side.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.