TOPICS
Search

Logarithmic Brunn-Minkowski Inequality


The logarithmic Brunn-Minkowski inequality compares the volume of the logarithmic interpolation of two convex bodies with the geometric mean of their volumes. For K,L subset R^n containing the origin in their interiors and 0<t<1, define

 C_t(K,L)={x in R^n:<x,u><=h_K(u)^(1-t)h_L(u)^t for all u in S^(n-1)},

where h_K(u)=max_(x in K)<x,u> and h_L are the support functions. The inequality is

 vol(C_t(K,L))>=vol(K)^(1-t)vol(L)^t.

Liu et al. (2026) reported a proof for all n>=2 when both centroids are at the origin, without assuming central symmetry. Equality holds iff there are complementary subspaces E_1, ..., E_r, centered convex bodies K_i subset E_i, and positive scalars a_i such that K=K_1 direct sum ... direct sum K_r and L=a_1K_1 direct sum ... direct sum a_rK_r. Here direct sums of sets are Minkowski sums in complementary subspaces, which need not be orthogonal.

Liu et al. (2026) state that all proofs were generated by ChatGPT and were subjected to adversarial checks. Independent verification of the reported proof is not asserted here.


See also

Brunn-Minkowski Inequality, Convex Body, Minkowski Sum, Support Function

Explore with Wolfram|Alpha

References

Liu, Y.; Sun, Q.; and Xiong, G. "The Centered Logarithmic Brunn-Minkowski Inequality." 7 Oct 2026. https://arxiv.org/abs/2610.10152.

Cite this as:

Weisstein, Eric W. "Logarithmic Brunn-Minkowski Inequality." From MathWorld--A Wolfram Resource. https://mathworld.wolfram.com/LogarithmicBrunn-MinkowskiInequality.html

Subject classifications