Exceptional set estimate through Brascamp-Lieb inequality
Shengwen Gan

TL;DR
This paper establishes new upper bounds for the dimension of exceptional sets under projections using Brascamp-Lieb inequalities, improving previous estimates and explicitly determining values for certain parameter ranges.
Contribution
It introduces a novel upper bound for the exceptional set dimension function T(a,s) utilizing Brascamp-Lieb inequality, refining earlier results and providing explicit calculations for specific cases.
Findings
New upper bound for T(a,s) using Brascamp-Lieb inequality
Improved estimate T(a,k/n a) ≤ k(n-k)-min{k,n-k}
Explicit values of T(a,s) for certain (a,s) ranges
Abstract
Fix integers , and numbers satisfying . The problem of exceptional set estimate is to determine \[T(a,s):=\sup_{A\subset \mathbb{R}^n\ \text{dim}A=a}\text{dim}(\{ V\in G(k,n): \text{dim}(\pi_V(A))<s \}). \] In this paper, we prove a new upper bound for by using Brascamp-Lieb inequality. As one of the corollary, we obtain the estimate \[T(a,\frac{k}{n}a)\le k(n-k)-\min\{k,n-k\}, \] which improves a previous result of He. By constructing examples, we can determine the explicit value of for certain : When , and , we have \[T(1+\beta,\gamma)=k(n-k)-k.\] When , and , we have \[T(n-1+\beta,k-1+\gamma)=k(n-k)-(n-k).\]
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
TopicsProbabilistic and Robust Engineering Design · Mathematical Approximation and Integration
