On Mahler’s conjecture for even s-concave functions in dimensions 1 and 2.

Matthieu Fradelizi, Elie Nakhle · 2024

Read On Mahler’s conjecture for even s-concave functions in dimensions 1 and 2. by Matthieu Fradelizi, Elie Nakhle (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-12372.txt.

Introduction

Matthieu Fradelizi and Elie Nakhle prove sharp forms of Mahler's conjecture for a family of even functions. Their 2024 paper works in dimensions one and two. The conjecture bounds a volume product pairing a function with its polar twin. A cube's indicator is the suspected minimizer.

The family is s-concave functions, wider than the log-concave case settled earlier. In dimension one, the proof now covers all negative s down to minus one. In dimension two, positive s values with whole reciprocals are settled fully. Negative s in two dimensions needs a natural extra condition on tails. The result extends a classic convex geometry program into new function classes.

Catalog record

CREATORMatthieu Fradelizi, Elie Nakhle
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE56 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-12372

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

On Mahler’s conjecture for even s-concave functions in dimensions 1 and 2. Matthieu Fradelizi, Elie Nakhle Abstract . In this paper, we establish different sharp forms of Mahler’s conjecture for s s -concave even functions in dimensions n n , for n = 1 n=1 and 2 2 , for s > − 1 / n s>-1/n , thus generalizing our previous results in [ FN ] on log-concave even functions in dimension 2, which corresponds to the case s = 0 s=0 . The functional volume product of an even s s -concave function g g is ∫ ℝ n g ⁡ ( x ) ​ 𝑑 x ​ ∫ ℝ n ℒ s ​ g ​ ( y ) ​ 𝑑 y , \int_{\mathbb{R}^{n}}g(x)dx\int_{\mathbb{R}^{n}}\mathcal{L}_{s}g(y)dy, where ℒ s ​ g \mathcal{L}_{s}g is the s s -polar function associated to g g . The analogue of Mahler’s conjecture for even s s -concave functions postulates that this quantity is minimized for the indicatrix of a cube for any s > − 1 / n s>-1/n . In dimension n = 1 n=1 , we prove this conjecture for all s ∈ ( − 1 , 0 ) s\in(-1,0) (the case s ≥ 0 s\geq 0 was established by the first author and Mathieu Meyer in [ FM10 , page 17] ). In dimension n = 2 n=2 , we consider the case 1 / s ∈ ℤ 1/s\in{\mathbb{Z}} : for s > 0 s>0 , we establish Mahler’s conjecture for general s s -concave even functions; for s < 0 s<0 , we prove a sharp inequality for s s -concave functions g g such that g s g^{s} admits an asymptote in every direction. Notice that this set of functions is quite natural to consider, when s < 0 s<0 , since it is the largest subset of s s -concave functions stable by s s -duality. 2020 Mathematics Subject Classification Primary 52A40, 52A20 1. Introduction. The classical Blaschke-Santaló inequality gives the following sharp relation between the volume of a centrally symmetric convex body K in \mathbb{R}^{n} and the volume of its polar body K^{\circ} : P(K)=|K||K^{\circ}|\leq P(B_{2}^{n}), where B_{2}^{n} is the Euclidean ball of radius one, K^{\circ}=\{y\in\mathbb{R}^{n},\langle x,y\rangle\leq 1,\forall x\in K\} is the polar body of K in \mathbb{R}^{n} and |A| stands for the Lebesgue measure of a Borel subset A of \mathbb{R}^{n} . One of the well-known problems in the field of geometry of convex bodies is an inverse form of the Blaschke-Santaló inequality, namely Mahler’s conjecture. This conjecture states that the product of the volume of a centrally symmetric convex body K and the volume of its polar body K^{\circ} , should be minimized by the cube, that is, P(K)\geq P([-1,1]^{n})=\frac{4^{n}}{n!}. This conjecture was stated in 1938 by Mahler [ Mah2 ] . The inequality in dimension two was established by Mahler himself [ Mah1 ] , the 3 -dimensional inequality and the equality case were proved by Iriyeh and Shibata [ IS ] , a shorter proof can be found in [ FHMRZ ] . The conjecture was also proved for several particular families of convex bodies like unconditional convex bodies by Saint Raymond [ SR , M ] , zonoids by Reisner [ Re , GMR ] , hyperplane sections of B_{p}^{n}=\{x\in\mathbb{R}^{n};\sum|x|_{i}^{p}\leq 1\} by Karasev [ K ] . Functional forms of Mahler’s conjecture have also been studied, where the centrally symmetric convex body is replaced with an even s -concave function and the polar body is replaced with the suitable s -polar of this function. More precisely, let us recall the definition of s -concavity. Definition 1.1 . Let s\in\mathbb{R} . A function g:\mathbb{R}^{n}\to\mathbb{R}_{+} is s -concave, if for every \lambda\in[0,1] , and every x , y\in\mathbb{R}^{n} such that g(x)g(y)>0 , one has g((1-\lambda)x+\lambda y)\geq((1-\lambda)g(x)^{s}+\lambda g(y)^{s})^{\frac{1}{s}},\quad\hbox{for}\ s\neq 0, and g((1-\lambda)x+\lambda y)\geq g(x)^{1-\lambda}g(y)^{\lambda} , for s=0 . Let g:\mathbb{R}^{n}\to\mathbb{R}_{+} be an s -concave function not identically zero. Notice that for s>0 , it means that g^{s} is concave on its support, while for s<0 , it means that g^{s}:\mathbb{R}^{n}\to\mathbb{R}_{+}\cup\{+\infty\} is convex. For s -concave functions, the s -duality takes the following form. We define, for every y\in\mathbb{R}^{n} , \mathcal{L}_{s}g(y):=\inf_{x\in\mathbb{R}^{n}}\frac{\left(1-s\langle x,y\rangle\right)_{+}^{\frac{1}{s}}}{g(x)},\quad\hbox{for}\ s\neq 0,\ \hbox{and}\quad\mathcal{L}_{0}g(y)=\inf_{x\in\mathbb{R}^{n}}\frac{e^{-\langle x,y\rangle}}{g(x)}, where the infimum are considered on the set \{x\in\mathbb{R}^{n};g(x)>0\} . Note that \mathcal{L}_{s}g is s -concave and that \mathcal{L}_{0} can also be described using the Legendre transform: for any function \varphi:\mathbb{R}^{n}\to\mathbb{R}\cup\{+\infty\} , one has \mathcal{L}_{0}(e^{-\varphi})=e^{-\varphi^{*}} , where \varphi^{*}(y)=\sup_{x}(\langle x,y\rangle-\varphi(x)) denotes the Legendre transform of \varphi . The following conjecture is the natural analogue of Mahler’s conjecture for even s -concave functions. Conjecture 1.2 . Let s>-\frac{1}{n} and g:\mathbb{R}^{n}\to\mathbb{R}_{+} be an even s -concave function such that 0<\int g<+\infty . Then, \int_{\mathbb{R}^{n}}g(x)dx\int_{\mathbb{R}^{n}}\mathcal{L}_{s}g(y)dy\geq\frac{4^{n}}{(1+s)\cdots(1+ns)}, with equality for g=\mathds{1}_{[-1,1]^{n}} . The conjecture has been proved for s=0 and g being unconditional in [ FM08a , FM08b ] . Recall that a function is unconditional if g(|x_{1}|,...,|x_{n}|)=g(x_{1},...,x_{n}) for every (x_{1},...,x_{n})\in\mathbb{R}^{n} . It has also been established for s>0 and n=1 in [ FM10 , page 17] ; for s>0 and g unconditional in [ BF , FGSZ ] and for s=0 and n=2 in [ FN ] . According to the sign of s , there are two ways of reformulating Conjecture 1.2 . If s>0 , then it is equivalent to finding the exact lower bound for the quantity \int_{\mathbb{R}^{n}}f(x)^{m}dx\int_{\mathbb{R}^{n}}(\mathcal{L}_{1}f(y))^{m}dy, where f is an even concave function such that 0<\int f^{m}<+\infty . While for s<0 , Conjecture 1.2 is equivalent to finding the exact lower bound for \int_{\mathbb{R}^{n}}\frac{1}{f(x)^{m}}dx\int_{\mathbb{R}^{n}}\frac{1}{(\mathcal{M}f(y))^{m}}dy, where f:\mathbb{R}^{n}\to(0,+\infty] is even, convex, f goes to infinity at infinity and for every y\in\mathbb{R}^{n} \mathcal{M}f(y)=\sup_{x\in\mathbb{R}^{n}}\frac{1+\langle x,y\rangle}{f(x)}. In our first main result, we prove Conjecture 1.2 in the case where 1/s\in\mathbb{N} , and in dimension n=2 . Theorem 1.3 . Let s>0 be such that 1/s\in\mathbb{N} and let g:\mathbb{R}^{2}\to\mathbb{R}_{+} be an even s -concave function such that 0<\int g<+\infty . Then, \int_{\mathbb{R}^{2}}g(x)dx\int_{\mathbb{R}^{2}}\mathcal{L}_{s}g(y)dy\geq\frac{16}{(1+s)(1+2s)}, with equality for g=\mathds{1}_{[-1,1]^{2}} . Notice that letting s\to 0 , we get a new proof of the log-concave case from [ FN ] , though without the equality case, see Corollary 3.3 . For s<0 , the situation is more involved, because the s -dual \mathcal{L}_{s}g of an integrable s -concave function g is not only s -concave but, as observed in [ Ro , FGSZ ] , the function (\mathcal{L}_{s}g)^{s} has also asymptotes in every direction. For example, for g=\mathds{1}_{[-1,1]^{n}} , one has (\mathcal{L}_{s}g(y))^{s}=1-s\|y\|_{1} , where \|y\|_{1}=\sum_{i=1}^{n}|y_{i}| , for y=(y_{1},\dots,y_{n}) . More precisely, in general, one has (\mathcal{L}_{s}g)^{s}\in\mathcal{F} , where the set \mathcal{F} is defined below. Definition 1.4 . \mathcal{F} is the set of convex lower semi-continuous functions f:\mathbb{R}^{n}\to(0,+\infty) such that, for any x\neq 0 , one has \lim_{t\to+\infty}f(tx)=+\infty , and t\mapsto\frac{f(tx)}{t} is non-increasing on (0,+\infty) . Notice that if g^{s}\in\mathcal{F} , then it is not difficult to see that 0<\int g<+\infty , see the inequality ( 2.3 ) in section 2. Moreover, it is easy to see that for any g such that g^{s}\in\mathcal{F} , one has \mathcal{L}_{s}\mathcal{L}_{s}g=g , from which it follows that for any map g , one has \mathcal{L}_{s}\mathcal{L}_{s}\mathcal{L}_{s}g=\mathcal{L}_{s}g . As we shall see from the one-dimensional case, it is natural to formulate two more conjectures from which Conjecture 1.2 follows in the case s<0 . The first

Read the complete text