OctoInformBeta

Logarithmic and Lp Brunn–Minkowski inequalities and the B-conjecture

A convex body in ℝn\mathbb{R}^n is a compact convex set with nonempty interior. A convex body KK is origin-symmetric when K=−KK=-K. The support function of a convex body KK is hK(u)=max⁡x∈K⟨x,u⟩h_K(u)=\max_{x\in K}\langle x,u\rangle, for uu in the unit sphere Sn−1S^{n-1}. The support function of an origin-symmetric convex body is positive. The Wulff body of a positive continuous function ff on Sn−1S^{n-1} is the set 𝒲[f]=⋂u∈Sn−1{x∈ℝn:⟨x,u⟩≤f(u)}\mathcal{W}[f]=\bigcap_{u\in S^{n-1}}\{x\in\mathbb{R}^n:\langle x,u\rangle\le f(u)\}. For origin-symmetric convex bodies KK and LL and 0≤λ≤10\le\lambda\le1, the logarithmic combination of KK and LL is 𝒲[hK1−λhLλ]\mathcal{W}[h_K^{1-\lambda}h_L^\lambda]. It is contained in the Minkowski combination (1−λ)K+λL(1-\lambda)K+\lambda L. Write |K||K| for the nn-dimensional Lebesgue measure of KK. In its multiplicative form, the Brunn-Minkowski inequality reads |(1−λ)K+λL|≥|K|1−λ|L|λ|(1-\lambda)K+\lambda L|\ge|K|^{1-\lambda}|L|^\lambda. The logarithmic Brunn-Minkowski conjecture asserts the same lower bound for the volume of the logarithmic combination, for all origin-symmetric convex bodies in every dimension.

The conjecture is due to Böröczky, Lutwak, Yang and Zhang. They proved it in the plane. Saroglou proved it for bodies unconditional with respect to the same orthonormal basis. Böröczky and Kalantzopoulos extended it to bodies invariant under the same nn linear reflections whose fixed hyperplanes meet only at the origin. Lu proved it for pairs of origin-symmetric bodies sharing n−2n-2 orthogonal reflection symmetries. For unit balls of complex norms, Rotem derived it from Cordero-Erausquin’s volume inequality for complex interpolation. A complementary approach studies the local form of the inequality, through the second variation of volume. Colesanti, Livshyts and Marsiglietti established a local result near Euclidean balls. Van Handel proved the local inequality for origin-symmetric zonoids against arbitrary origin-symmetric comparison bodies. Stancu announced the inequality in all dimensions, in a report that deferred some details.

The preprint proves the conjecture: for n≥1n\ge1, origin-symmetric convex bodies K,L⊂ℝnK,L\subset\mathbb{R}^n and 0≤λ≤10\le\lambda\le1, |𝒲[hK1−λhLλ]|≥|K|1−λ|L|λ|\mathcal{W}[h_K^{1-\lambda}h_L^\lambda]|\ge|K|^{1-\lambda}|L|^\lambda. For 0<p<10<p<1, the LpL_p combination of origin-symmetric convex bodies KK and LL is (1−λ)⋅K+pλ⋅L=𝒲[((1−λ)hKp+λhLp)1/p](1-\lambda)\cdot K+_p\lambda\cdot L=\mathcal{W}[((1-\lambda)h_K^p+\lambda h_L^p)^{1/p}]. The LpL_p Brunn-Minkowski inequality reads |(1−λ)⋅K+pλ⋅L|p/n≥(1−λ)|K|p/n+λ|L|p/n|(1-\lambda)\cdot K+_p\lambda\cdot L|^{p/n}\ge(1-\lambda)|K|^{p/n}+\lambda|L|^{p/n}. From the conjecture the preprint derives the LpL_p Brunn-Minkowski inequality for origin-symmetric convex bodies, for every 0<p<10<p<1. With Saroglou’s transfer theorem and a reduction for measures supported on proper subspaces, the conjecture yields its inequality for every even log-concave measure μ\mu in place of volume, μ(𝒲[hK1−λhLλ])≥μ(K)1−λμ(L)λ\mu(\mathcal{W}[h_K^{1-\lambda}h_L^\lambda])\ge\mu(K)^{1-\lambda}\mu(L)^\lambda. The preprint also obtains the (B)(B)-conjecture that t↦μ(etK)t\mapsto\mu(e^tK) is log-concave for every such measure μ\mu and every origin-symmetric convex body KK. The Lean library contains none of these.

The Lean library LogVolume proves this theorem as the preprint states it, with no further hypothesis. Lean has checked every result of the Lean library, so the Lean library is the reference. Each result on these pages is stated as the Lean library states it and linked to the Lean declaration behind it, with the preprint’s statement beside it where the two differ. The preprint is a guide only: it gives the results their order and their names.

The Lean library: LogVolume

LogVolume is a Lean library: 16 items, in seven groups. Each has a page written by OctoInform, listed below.

§1. Introduction 2 items
  1. Convex bodies, support functions and Wulff bodies Logarithmic combination of convex bodies through support functions.
  2. Theorem 1.1 Origin-symmetric convex bodies satisfy the logarithmic Brunn-Minkowski inequality.
§2. From a variance bound to volume 5 items
  1. The soft slab potential and its law (2.1) The soft slab potential and its Gibbs law.
  2. Proposition 2.1 A variance bound for the soft slab law.
  3. Log-concavity of the partition function (2.3) A variance bound makes the partition function log-concave.
  4. Finitely many slabs (2.4) From soft slab integrals to the volume of slab bodies.
  5. All directions and dimension one The log-Brunn-Minkowski inequality from slab bodies and a variance bound.
§3. Moment coordinates on the whole space 1 item
  1. Lemma 3.1 Soft slab laws are moment measures of smooth potentials.
§4. Adjoints in moment coordinates 3 items
  1. τ, the operator A, and the tests W and u (4.2), (4.4) The moment operator and its test-function fields in target coordinates.
  2. The integration formula (4.3) and the gradient identity (4.6) Integration by parts and flux identities for the moment operator.
  3. The Hessian identity (4.7) The mean Hessian energy in terms of the moment operator.
§5. Density of the even tests 1 item
  1. Lemma 5.1 Centered even functions are limits of A(A−1)fA(A-1)f in L2(μ)L^2(\mu).
§6. A tensor estimate for the divergence tests 1 item
  1. Lemma 6.1 Tensor estimates for a convex function and its moment measure.
§7. The variance inequality (proof of Proposition 2.1) 3 items
  1. Homogeneity and the trace moment (7.1), (7.2) Euler relations for slab powers and the trace moment identity.
  2. The pairings and the two Cauchy-Schwarz bounds (7.3) to (7.5) Three forms of one pairing and two Cauchy-Schwarz bounds.
  3. The dual bound (7.6), density, and the conclusion The dual bound and the soft slab variance inequality.

No item matches.