Friedgut–Kahn–Kalai–Keller spectral correlation conjecture

Less than 1 year old · traced to

We work on the discrete cube {0,1}n\{0,1\}^n with the uniform product measure. For functions f,gf,g on the cube, write

Cov⁡(f,g)=E[fg]−E[f]E[g].\operatorname{Cov}(f,g)=\mathbb{E}[fg]-\mathbb{E}[f]\mathbb{E}[g].

A function is increasing if x≤yx\leq y coordinatewise implies f(x)≤f(y)f(x)\leq f(y). For a Boolean function ff, let f^(S)\hat f(S) denote its Fourier coefficient on S⊆[n]S\subseteq[n].

Friedgut–Kahn–Kalai–Keller spectral correlation conjecture. For any increasing Boolean functions f,g:{0,1}n→{0,1}f,g:\{0,1\}^n\to\{0,1\},

Cov⁡(f,g)≥4∑S≠∅∣S∣f^(S)2g^(S)2.\operatorname{Cov}(f,g)\geq 4\sum_{S\neq\emptyset}|S|\hat f(S)^2\hat g(S)^2.

This conjecture is a spectral strengthening of the Harris–Kleitman inequality and would sharpen Talagrand’s correlation theory. The paper proves the same inequality with constant 22, while the conjectured constant 44 is not resolved here; related structured cases were previously verified by Chang and Chen.

References

Primary source

Fan Chang, “A spectral correlation inequality for increasing Boolean functions”, arXiv:2606.08958 (2026).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.