Dimension Reduction in $L_p$, $0<p<2$
Gideon Schechtman

TL;DR
This paper demonstrates that for all 0<p<2, any set of k points in L_p space can be embedded into finite-dimensional l_p space with arbitrarily small distortion, with dimension linear in k and polynomial in 1/ε.
Contribution
It extends previous results by showing that for all 0<p<2, embeddings with low distortion are possible into finite-dimensional l_p spaces with dimension linear in the number of points.
Findings
Embedding dimension is linear in the number of points k.
Distortion can be made arbitrarily close to 1.
Applicable for all 0<p<2, generalizing prior work for p=1.
Abstract
Complementing a recent observation of Newman and Rabinovich for we observe here that for all any points in embeds with distortion into where is linear in (and polynomial in ).
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsMathematical Analysis and Transform Methods · Algebraic and Geometric Analysis · Advanced Harmonic Analysis Research
