2016
DOI: 10.1017/fmp.2016.1
|View full text |Cite
|
Sign up to set email alerts
|

Metric Inequalities

Abstract: For every p ∈ (0, ∞) we associate to every metric space (X, d X ) a numerical invariantThus, our new bi-Lipschitz invariant certifies that L q does not admit a bi-Lipschitz embedding into L p when 2 < q < p < ∞. This completes the long-standing search for bi-Lipschitz invariants that serve as an obstruction to the embeddability of L p spaces into each other, the previously understood cases of which were metric notions of type and cotype, which however fail to certify the nonembeddability of L q into L p when 2… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

2
29
0

Year Published

2016
2016
2021
2021

Publication Types

Select...
8
1

Relationship

5
4

Authors

Journals

citations
Cited by 13 publications
(31 citation statements)
references
References 88 publications
(207 reference statements)
2
29
0
Order By: Relevance
“…In particular, the metric spaces {scriptQp:1<p2} are mutually nonbilipschitz homeomorphic. The cases where q>max{p,2} or 2<q<p (in which it is known that Lq does not bilipschitz embed in Lp) are more subtle, as seeking for metric invariants which would explain the corresponding nonembeddability results for Lp‐spaces proved to be a very difficult problem (see , and the references therein).…”
Section: Examplesmentioning
confidence: 99%
“…In particular, the metric spaces {scriptQp:1<p2} are mutually nonbilipschitz homeomorphic. The cases where q>max{p,2} or 2<q<p (in which it is known that Lq does not bilipschitz embed in Lp) are more subtle, as seeking for metric invariants which would explain the corresponding nonembeddability results for Lp‐spaces proved to be a very difficult problem (see , and the references therein).…”
Section: Examplesmentioning
confidence: 99%
“…The above discretization problem was introduced in [13] as an alternative (quantitative) approach to an important rigidity theorem of Ribe [82]. Additional applications to embedding theory appear in [68,39,69]. To date, the bound (2) remains the best known, even under the additional restriction that Y is uniformly convex.…”
Section: Amentioning
confidence: 99%
“…Beyond metric cotype, dimension-dependent scaling parameters occur (for conceptually distinct reasons) in other metric inequalities that arise in the Ribe program. Determining their asymptotically sharp values is a major difficulty that pertains to important open problems; see [120,61,176,60,135,130,137,48]. The currently best-known general bound [60] for the metric cotype q scaling parameter of Banach spaces of Rademacher cotype q is m n 1+1/q .…”
Section: Theorem 2 There Exists a Metric Space Z That Does Not Embedmentioning
confidence: 99%