Kolmogorov rearrangement problem and Garsia's conjecture

Let (X,μ)(X,\mu) be a probability space and let (φn)n≥1(\varphi_n)_{n\ge 1} be a complete uniformly bounded orthonormal system in L2(X,μ)L^2(X,\mu), meaning that sup⁡n∥φn∥∞<∞\sup_n\lVert\varphi_n\rVert_{\infty}<\infty. The rearrangement assertion asks whether there always exists a permutation π\pi of N\mathbb{N} such that, for every coefficient sequence (an)∈ℓ2(a_n)\in\ell^2, the series ∑n=1∞anφπ(n)(x)\sum_{n=1}^{\infty}a_n\varphi_{\pi(n)}(x) converges for μ\mu-almost every x∈Xx\in X. Equivalently, the claimed counterexample is a complete uniformly bounded orthonormal system for which, for every permutation π\pi, there exists (an)∈ℓ2(a_n)\in\ell^2 such that ∑n=1∞anφπ(n)(x)\sum_{n=1}^{\infty}a_n\varphi_{\pi(n)}(x) diverges for almost every xx.

Equivalent formulations 1Other wordings

Other statements of this same problem, merged from separate entries. Each is equivalent to the statement above — proving any one settles them all.

  1. Counterexample formulation

    There exists a complete uniformly bounded orthonormal system (φn)(\varphi_n) such that for every permutation π\pi of N\mathbb{N}, there is a sequence (an)∈ℓ2(a_n)\in\ell^2 for which the rearranged orthogonal series ∑n=1∞anφπ(n)\sum_{n=1}^{\infty}a_n\varphi_{\pi(n)} diverges almost everywhere.

    source: On Kolmogorov's rearrangement problem and Garsia's conjecture

References

Primary source

arXiv

Additional references

Progress summary

Refreshed
Claimed solved

A September 2026 preprint claims to disprove both longstanding questions, but its result has not yet been independently confirmed.

The problem concerns two longstanding questions about rearrangements of orthonormal systems and almost-everywhere convergence. The claimed construction uses two differently ordered copies of the trigonometric system to give negative answers to both questions.

September 2026 claimed counterexample

Mark Lewko's preprint claims a construction that disproves both the Kolmogorov rearrangement assertion and Garsia's conjecture. The claim is explicitly awaiting peer review, so it remains unverified.

Current status (as of September 2026): Both questions are claimed to have negative answers, but the construction remains unverified pending peer review.

Sources

Solutions 0

No solutions have been posted yet.