← All problems

The spherical-cap simplex conjecture

9 years old · traced to

Let Sn−1\mathbb S^{n-1} be the unit sphere, and consider N=n+1N=n+1 spherical caps of radius RR on it. The spherical-cap simplex conjecture. For every R>0R>0, the maximal area of the union of these NN caps is attained when their centers form a regular simplex inscribed into the unit sphere. This is presented as an even stronger statement implying the union-of-balls conjecture.

References
  1. Alexey Balitskiy, Roman Karasev and Alexander Tsigler, “Optimality of codes with respect to error probability in Gaussian noise”, arXiv:1701.07986 (2017).

Progress summary

Refreshed
Open · progress reported

The conjecture remains open in higher dimensions; a 2026 paper proves only an averaged variant, not the original claim.

The conjecture says that n+1n+1 equal spherical caps on Sn−1\mathbb S^{n-1} have largest union when their centers form a regular simplex. The general statement remains unresolved; an earlier attempted proof had a gap in higher dimensions.

Known results

  • The n=3n=3 case follows from the two-dimensional-sphere result of Landau–Slepian (1966).
  • In an equivalent indexing, the problem is elementary for n≤3n\leq 3 and affirmative for n=4n=4.
  • The nontrivial range is 1n−1<c<n−22(n−1)\frac{1}{n-1}<c<\sqrt{\frac{n-2}{2(n-1)}}; the case n≥5n\geq 5 remains open.

2026 averaged result

A 2026 arXiv paper proves that the regular simplex is optimal after averaging the spherical-cap union measure over a random cap radius. This is a genuine partial result, but it does not prove the pointwise conjecture; the claim is reported here without independent verification.

Current status (as of November 2026): The low-dimensional cases recorded above are settled, while the original pointwise conjecture for n≥5n\geq 5 remains open; only an unverified averaged-radius advance is reported.

Sources

Solutions 0

No solutions have been posted yet.