Does \ell^1 have metric-Markov-cotype-2?
Does \ell^1 have metric-Markov-cotype-2?
References
K. Ball, Markov chains, Riesz transforms and Lipschitz maps, Geom. Funct. Anal. 2 (1992), no. 2, 137–172.
Progress summary
A reader-submitted proof claims a positive answer, but no independent source has verified it.
The problem asks whether has metric Markov cotype . Published work records this as an open question and notes that a positive answer would resolve a longstanding Lipschitz-extension problem attributed to Ball.
Known results
- Some closed subspaces of fail to have metric Markov cotype; this does not settle itself.
- Metric Markov cotype for would imply finiteness of .
September 10, 2026 community submission
A submitted proof argues that real and complex spaces have metric Markov cotype , with explicit universal constants. The argument is unverified and has no independently checkable supporting source.
Current status (as of September 2026): The question remains open in the published record, while a community submission claims a positive answer that is unverified.
Sources
Solutions 2
We prove that real and complex L1 spaces have metric Markov cotype 2 with explicit universal constants.See full solution
- Metric_Markov_Cotype_L1_Polished.pdfOpen
Claimed by OpenAI. Claims metric Markov cotype two for real l_1 and consequently universal-loss Lipschitz extension into l_1 from arbitrary subsets of real Hilbert spaces.See full solution
Scope relative to this problem: The source claims metric Markov cotype 2 for real l_1 and, as a separate consequence, universal-loss Lipschitz extension into real l_1 from arbitrary subsets of real Hilbert spaces. The target metric property is directly addressed; no linear extension operator or extension to arbitrary other Banach targets is asserted.
GitHub repository: https://github.com/openai/math
- OpenAI-332-01-Metric-Markov-Cotype-Two-of-1.pdfOpen