TOPICS
Search

Autoconvolution Supremum Problem


The autoconvolution supremum problem asks how small the essential supremum of the convolution of a nonnegative function with itself can be when its integral and the length of a containing interval are fixed (Martin and O'Bryant 2009, Matolcsi and Vinuesa 2010). Let F consist of the nonnegative integrable functions supported on [-1/4,1/4] with integral 1. The problem is to determine

 S=inf_(f in F)||f*f||_infty,
(1)

where f*f denotes the convolution of f(x) with itself and ||·||_infty is the essential supremum norm. Equivalently, S is the largest constant for which

 ||f*f||_infty>=S(int_Rf(x)dx)^2
(2)

for every nonzero nonnegative integrable function f(x) supported on this interval. With support [0,1] instead, the corresponding optimal constant is S/2 by a change of scale.

In probabilistic terms, [f*f](t) is the probability density function of X+Y for f in F, when X and Y are independent and identically distributed with probability density function f(x) (Matolcsi and Vinuesa 2010).

An explicit step function gives an upper bound for S. If f(x) has nonnegative heights a_0, a_1, ..., a_(n-1), not all zero, on n equal subintervals of [-1/4,1/4], then

 (||f*f||_infty)/((int_Rf(x)dx)^2)=(2nmax_(0<=k<=2n-2)sum_(i=0)^(n-1)a_ia_(k-i))/((sum_(i=0)^(n-1)a_i)^2),
(3)

where a_j=0 outside 0<=j<n. The maximum can be evaluated at the breakpoints because the convolution is a piecewise linear function (Martin and O'Bryant 2009, Matolcsi and Vinuesa 2010).

Yuksekgonul et al. (2026) used AI-assisted search to find a 30,000-piece step function giving

 S<1.50287.
(4)

The autoconvolution inequality concerns a different ratio of norms, without this fixed-support restriction.

Ivanisvili and Xie (2026) refined the discretization estimate of Cloninger and Steinerberger (2017) to prove

 S>=1.2802.
(5)

They report that Grok suggested the refinement. Atris Labs (2026) constructed a 65,536-piece step function with rational heights giving

 S<1.5027434868.
(6)

An improved lower bound for S also gives an improved upper bound for the size of a generalized Sidon set.

The exact value remains unknown.


See also

Autoconvolution Inequality, Convolution, Step Function

Explore with Wolfram|Alpha

References

Atris Labs. "Atris Records." Oct. 13, 2026. https://github.com/atrislabs/records/tree/1fa857a021ddbb4064708daf336758b866f75cb4.Cloninger, A. and Steinerberger, S. "On Suprema of Autoconvolutions with an Application to Sidon Sets." Proc. Amer. Math. Soc. 145, 3191-3200, 2017. https://doi.org/10.1090/proc/13690.Ivanisvili, P. and Xie, X. "Grokability in Five Inequalities." 6 May 2026. https://arxiv.org/abs/2605.05193v1.Martin, G. and O'Bryant, K. "The Supremum of Autoconvolutions, with Applications to Additive Number Theory." Illinois J. Math. 53, 219-235, 2009. https://doi.org/10.1215/ijm/1264170847.Matolcsi, M. and Vinuesa, C. "Improved Bounds on the Supremum of Autoconvolutions." J. Math. Anal. Appl. 372, 439-447.e2, 2010. https://doi.org/10.1016/j.jmaa.2010.07.030.Yuksekgonul, M.; Koceja, D.; Li, X.; Bianchi, F.; McCaleb, J.; Wang, X.; Kautz, J.; Choi, Y.; Zou, J.; Guestrin, C.; and Sun, Y. "Learning to Discover at Test Time." 5 Feb 2026. https://arxiv.org/abs/2601.16175v2.

Cite this as:

Weisstein, Eric W. "Autoconvolution Supremum Problem." From MathWorld--A Wolfram Resource. https://mathworld.wolfram.com/AutoconvolutionSupremumProblem.html

Subject classifications