The log-Brunn–Minkowski conjecture

About 7 years old · traced to

Let KK and LL be convex bodies in Rn\mathbb{R}^n. For λ∈(0,1)\lambda\in(0,1), their geometric mean is defined, for origin-symmetric bodies, by

λ⋅K+0(1−λ)⋅L:={x∈Rn:⟨x,u⟩≤hKλ(u)hL1−λ(u) ∀u∈Sn−1}.\lambda\cdot K+_0(1-\lambda)\cdot L:=\{x\in\mathbb{R}^n:\langle x,u\rangle\leq h_K^{\lambda}(u)h_L^{1-\lambda}(u)\ \forall u\in S^{n-1}\}.

Here VV denotes volume, int⁡\operatorname{int} denotes interior, and oo is the origin. Log-Brunn–Minkowski conjecture. For any pair KK and LL of convex bodies in Rn\mathbb{R}^n, there exist zK∈int⁡Kz_K\in\operatorname{int}K and zL∈int⁡Lz_L\in\operatorname{int}L such that, for every λ∈(0,1)\lambda\in(0,1),

V((1−λ)⋅(K−zK)+0λ⋅(L−zL))≥V(K)1−λV(L)λ.V\bigl((1-\lambda)\cdot(K-z_K)+_0\lambda\cdot(L-z_L)\bigr)\geq V(K)^{1-\lambda}V(L)^\lambda.

If KK and LL are origin symmetric, then zK=zL=oz_K=z_L=o. Moreover, equality holds if and only if

K=K1+⋯+Km,L=L1+⋯+Lm,K=K_1+\cdots+K_m,\qquad L=L_1+\cdots+L_m,

for compact convex sets K1,…,Km,L1,…,LmK_1,\ldots,K_m,L_1,\ldots,L_m of dimension at least one, with ∑i=1mdim⁡Ki=n\sum_{i=1}^m\dim K_i=n, and KiK_i and LiL_i homothetic for i=1,…,mi=1,\ldots,m.

References

Primary source

Károly J. Böröczky and Pavlos Kalantzopoulos, “Log-Brunn-Minkowski inequality under symmetry”, arXiv:2002.12239 (2022).

Additional references

2 papers in this index state this conjecture (2019–2020). The statement above is taken from the most recent of them; the others are arXiv:1909.03729.

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.