← All problems

Bishop’s conjecture on the dimension of harmonic measure

For every integer n≥3n\ge 3, the sharp universal upper bound for the Hausdorff dimension of harmonic measure on domains in Rn\mathbb{R}^n is n−1n−1n-\frac{1}{n-1}. Equivalently, for every domain Ω⊂Rn\Omega\subset\mathbb{R}^n and every pole x∈Ωx\in\Omega, harmonic measure ωΩx\omega^x_\Omega satisfies dim⁡H(ωΩx)≤n−1n−1\dim_H(\omega^x_\Omega)\le n-\frac{1}{n-1}, and this bound is sharp.

References

Source: arXiv

Progress summary

Refreshed
Open · progress reported

A new unrefereed paper substantially improves numerical bounds toward Bishop’s conjecture, but does not settle it.

Bishop conjectured in 1992 that the sharp deficit in the dimension of harmonic measure is bn=1/(n−1)b_n=1/(n-1) in dimensions n≥3n\ge 3. The conjecture remains unresolved.

Known results

  • Bourgain, 1987: bn>0b_n>0 for n≥3n\ge 3.
  • Jones and Wolff, 1988: b2=1b_2=1.
  • Wolff, 1995: examples show bn<1b_n<1.
  • Badger and Genschaw, 2022: bn≥c n−2n(n−1)/ln⁡(n)b_n\ge c\,n^{-2n(n-1)}/\ln(n), including b3≥10−15b_3\ge 10^{-15} and b4≥2×10−26b_4\ge 2\times 10^{-26}.

October 2026 quantitative improvement

Michael Albert, Matthew Badger, and Alyssa Genschaw report new one-ring bounds giving deficits of 10−510^{-5} in dimension 33 and 10−710^{-7} in dimension 44, together with an improved general-dimensional estimate. The claims are unrefereed and constitute progress toward, not a proof of, Bishop’s conjecture.

Current status (as of October 2026): The planar case is settled, while Bishop’s sharp conjecture in dimensions n≥3n\ge 3 remains open; new quantitative bounds are claimed but unverified.

Sources

Solutions 0

No solutions have been posted yet.