Algorithmic Information Bounds for Distances and Orthogonal Projections

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cholak, Peter, Csörnyei, Marianna, Lutz, Neil, Lutz, Patrick, Mayordomo, Elvira, Stull, D. M.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912573399498752
author Cholak, Peter
Csörnyei, Marianna
Lutz, Neil
Lutz, Patrick
Mayordomo, Elvira
Stull, D. M.
author_facet Cholak, Peter
Csörnyei, Marianna
Lutz, Neil
Lutz, Patrick
Mayordomo, Elvira
Stull, D. M.
contents We develop quantitative algorithmic information bounds for orthogonal projections and distances in the plane. Under mild independence conditions, the distance $|x-y|$ and a projection coordinate $p_e x$ each retain at least half the algorithmic information content of $x$ in the sense of finite-precision Kolmogorov complexity, up to lower-order terms. Our bounds support conditioning on coarser approximations, enabling case analyses across precision scales. The proofs introduce a surrogate point selection step. Via the point-to-set principle we derive a new bound on the Hausdorff dimension of pinned distance sets, showing that every analytic set $E\subseteq\mathbb{R}^2$ with $\dim_H(E)\leq 1$ satisfies \[\sup_{x\in E}\dim_H(Δ_x E)\geq \frac{3}{4}\dim_H(E).\] We also extend Bourgain's theorem on exceptional sets for orthogonal projections to all sets that admit optimal Hausdorff oracles.
format Preprint
id arxiv_https___arxiv_org_abs_2509_05211
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Algorithmic Information Bounds for Distances and Orthogonal Projections
Cholak, Peter
Csörnyei, Marianna
Lutz, Neil
Lutz, Patrick
Mayordomo, Elvira
Stull, D. M.
Computational Complexity
Classical Analysis and ODEs
We develop quantitative algorithmic information bounds for orthogonal projections and distances in the plane. Under mild independence conditions, the distance $|x-y|$ and a projection coordinate $p_e x$ each retain at least half the algorithmic information content of $x$ in the sense of finite-precision Kolmogorov complexity, up to lower-order terms. Our bounds support conditioning on coarser approximations, enabling case analyses across precision scales. The proofs introduce a surrogate point selection step. Via the point-to-set principle we derive a new bound on the Hausdorff dimension of pinned distance sets, showing that every analytic set $E\subseteq\mathbb{R}^2$ with $\dim_H(E)\leq 1$ satisfies \[\sup_{x\in E}\dim_H(Δ_x E)\geq \frac{3}{4}\dim_H(E).\] We also extend Bourgain's theorem on exceptional sets for orthogonal projections to all sets that admit optimal Hausdorff oracles.
title Algorithmic Information Bounds for Distances and Orthogonal Projections
topic Computational Complexity
Classical Analysis and ODEs
url https://arxiv.org/abs/2509.05211