← All problems

Parcet–Randrianantoanina 2-co-lacunary subsequence problem

For every von Neumann algebra M\mathcal{M}, every bounded sequence (xn)n≥1(x_n)_{n\geq 1} in the predual L1(M)=M∗L_1(\mathcal{M})=\mathcal{M}_* has a subsequence that either converges in norm or is 22-co-lacunary; that is, ∀M ∀(xn)⊆L1(M)\forall \mathcal{M}\ \forall (x_n)\subseteq L_1(\mathcal{M}), sup⁡n∥xn∥<∞\sup_n\|x_n\|<\infty implies that there exists a subsequence (xnk)(x_{n_k}) which is norm-convergent or 22-co-lacunary.

References

Progress summary

Refreshed
Open · solution reported

A new preprint claims to settle the problem in full generality, but its proof has not been independently checked.

The problem asks whether every bounded sequence in the predual of a von Neumann algebra has either a norm-convergent subsequence or a 22-co-lacunary subsequence. It was posed by Parcet and Randrianantoanina.

Known results

  • For hyperfinite and semifinite von Neumann algebras, the dichotomy was established earlier, cited as Parcet–Randrianantoanina, Corollary 3.7.

October 13, 2026 claimed solution

Jinghao Huang and Yunpeng Zhu claim that for every von Neumann algebra M\mathcal{M}, every bounded sequence in L1(M)=M∗L_1(\mathcal{M})=\mathcal{M}_* has either a norm-convergent subsequence or a 22-co-lacunary subsequence, removing both prior assumptions. This is an unrefereed preprint claim, with no independent verification found.

Current status (as of October 2026): The hyperfinite and semifinite cases are known, while Huang and Zhu claim the full arbitrary-von-Neumann-algebra statement; that claimed resolution remains unverified.

Sources

Solutions 0

No solutions have been posted yet.