Talagrand's operator cotype problem

For each q≥2q\ge 2, does there exist a universal constant Cq<∞C_q<\infty such that every bounded linear operator T:X→YT:X\to Y between Banach spaces satisfies

Cot⁡Rad,q(T)≤Cqmax⁡{Cot⁡Gauss,q(T), πq,1(T)},\operatorname{Cot}_{\mathrm{Rad},q}(T)\le C_q\max\left\{\operatorname{Cot}_{\mathrm{Gauss},q}(T),\,\pi_{q,1}(T)\right\},

where Cot⁡Rad,q(T)\operatorname{Cot}_{\mathrm{Rad},q}(T) and Cot⁡Gauss,q(T)\operatorname{Cot}_{\mathrm{Gauss},q}(T) denote the operator's Rademacher and Gaussian cotype constants, respectively, and πq,1(T)\pi_{q,1}(T) is its (q,1)(q,1)-summing norm?

References

Primary source

arXiv

Additional references

Progress summary

Refreshed
Claimed solved

An unrefereed preprint claims a counterexample that disproves the proposed universal inequality in the case q=2q=2.

Talagrand's operator cotype problem concerns a proposed universal inequality for operators. No proposer or original date is identified in the retrieved material.

September 17, 2026 counterexample

Xinglong Wu's preprint A Counterexample to Talagrand's Operator Cotype Problem claims to disprove the universal inequality for q=2q=2. If correct, this reverses the proposed general principle; the preprint is unrefereed and the claim is therefore unverified.

Current status (as of September 2026): The universal inequality is claimed false for q=2q=2, but the counterexample remains unverified; the retrieved evidence does not settle other values of qq.

  • ChatGPT (GPT-5.6)OpenAIsolved2026-09-17evidence

    Talagrand’s operator cotype problem receives a counterexample

Sources

Solutions 0

No solutions have been posted yet.