Optimal Poincaré-Hardy-type Inequalities on Manifolds and Graphs
Florian Fischer and Christian Rose · 2025
Read Optimal Poincaré-Hardy-type Inequalities on Manifolds and Graphs by Florian Fischer and Christian Rose (2025) 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-2501-18379.txt.
Introduction
Florian Fischer and Christian Rose review and extend a method for sharp inequalities on curved spaces and on graphs, in a 2025 paper. The inequalities, of Poincare and Hardy type, bound how a function can spread given its slope, and the best possible weights in them are the prize. The method is first recalled for hyperbolic spaces.
The review then covers manifolds with like traits. A known result for regular trees, which are graphs that branch evenly, gets a new proof by the same method. New results follow for graphs with a weak spherical symmetry, including fast-growing trees and anti-trees. Their optimal weights grow larger at infinity than the classic weights built by an older recipe. The paper is a technical survey and advance for specialists in analysis on spaces and networks.
Catalog record
| CREATOR | Florian Fischer and Christian Rose |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 66 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-18379
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Optimal Poincaré-Hardy-type Inequalities on Manifolds and Graphs Florian Fischer and Christian Rose Abstract . We review a method to obtain optimal Poincaré-Hardy-type inequalities on the hyperbolic spaces, and discuss briefly generalisations to certain classes of Riemannian manifolds. Afterwards, we recall a corresponding result on homogeneous regular trees and provide a new proof using the aforementioned method. The same strategy will then be applied to obtain new optimal Hardy-type inequalities on weakly spherically symmetric graphs which include fast enough growing trees and anti-trees. In particular, this yields optimal weights which are larger at infinity than the optimal weights classically constructed via the Fitzsimmons ratio of the square root of the minimal positive Green’s function. 2020 Mathematics Subject Classification. Primary 35B09 ; Secondary 26D10; 31C12; 31C20; 35B09; 39A12; 58J60. Keywords. Poincaré-Hardy-type inequalities, hyperbolic spaces, Damek-Ricci spaces, homogeneous trees, weakly spherically symmetric graphs 1. Introduction A century ago, in the attempt of finding a simple and elegant proof of Hilbert’s double series inequality, Hardy proved the following inequality on the standard line graph \mathbb{N}_{0} : For all \varphi\colon\mathbb{N}_{0}\to\mathbb{R} with \varphi(0)=0 , we have \sum_{n=0}^{\infty}\left\lvert\varphi(n)-\varphi(n+1)\right\rvert^{2}\geq\frac{1}{4}\sum_{n=1}^{\infty}\frac{\varphi^{2}(n)}{n^{2}}. Soon afterwards, he also proved its counterpart in the continuum, i.e., for all smooth and compactly supported functions \varphi\colon(0,\infty)\to\mathbb{R} we have \int_{0}^{\infty}\left\lvert\varphi^{\prime}(x)\right\rvert^{2}\,\mathrm{d}x\geq\frac{1}{4}\int_{0}^{\infty}\frac{\varphi^{2}(x)}{x^{2}}\,\mathrm{d}x. Because of Hardy’s contributions, inequalities of this kind – in the discrete or continuum – nowadays bear his name. Versions of Hardy’s inequality to d -dimensional Euclidean space and large classes of Riemannian manifolds including Cartan-Hadamard manifolds [ BEL15 , Car97 ] were the starting point of diverse generalizations to various contexts (see e.g. [ Dav99 , DD14 , Gri99 , KMP06 , KO09 , KO13 , RS19 , Sch22 , Stu24 , TU23 ] ). In this paper, we focus on certain classes of smooth Riemannian manifolds and locally finite graphs. Generally speaking, on a measure space (X,m) , a Hardy inequality with respect to a quadratic form E holds on some reasonable large subset F of L^{2}(X,m) with norm \left\lVert\cdot\right\rVert if there is a non-trivial weight function w such that E(f)\geq\left\lVert f\right\rVert^{2}_{w},\qquad f\in F, where \left\lVert\cdot\right\rVert_{w}:=\left\lVert\cdot\sqrt{w}\right\rVert . The weight w is then called a Hardy weight . If (X,m) is a measure space equipped with a (pseudo-)distance d , then the classical Hardy weight is given by w=Cd^{-2}(o,\cdot) \text{for some }C>0\text{ and }o\in X . The optimal constant C>0 in the classical Hardy weight depends on the geometric properties of the underlying space. By now classical are the cases \mathbb{R}^{d} with constant (d-2)^{2}/4 , d\geq 3 , while for the real hyperbolic space \mathbb{H}^{d} , it is (d-1)^{2}/4 , d\geq 2 . For further optimal constants for special subdomains or other spaces, see e.g. [ Bar24 , DP23 , MMP98 ] and references therein. If the weight is chosen to be the constant function w=\lambda>0 , i.e., E(f)\geq\lambda\left\lVert f\right\rVert^{2},\qquad f\in F, the corresponding Hardy inequality is referred to as Poincaré inequality or L^{2} -gap. The latter comes from the fact that the best constant in the Poincaré inequality is given by the bottom of the L^{2} -spectrum of the generator of E . The Poincaré inequality usually requires stronger assumptions than the classical Hardy inequality, e.g. on Cartan-Hardamard manifolds the sectional curvature needs to be strictly negative. Our main motivation is to show the existence of spaces where the classical Hardy and Poincaré inequalities can be combined to Poincaré-Hardy-type inequalities . To be more precise, we find spaces for which E(f)\geq\|f\|^{2}_{w},\qquad f\in F, is optimal where w=\lambda+Cd^{-2}(o,\cdot)+\tilde{w} for o\in X , some constants \lambda,C>0 and function \tilde{w}\geq 0 . The aim of this paper is twofold: Firstly, we will review a simple method to get optimal Hardy-type inequalities on special Cartan-Hadamard manifolds. In cases where we have specific knowledge about the volume density function and the bottom of the spectrum, it will turn out that we get optimal Poincaré-Hardy inequalities. We discuss the method exemplarily and in detail for the hyperbolic space and then review its generalisations and extensions briefly. Secondly, we will focus on the discrete counterpart. So far, only results about Poincaré-Hardy-type inequalities on trees are known, [ BSV21 ] . We show how these results can be generalised to the much larger class of weakly spherically symmetric graphs. Moreover, we provide explicit calculations of the optimal Hardy weights with our new method. This is a simplification in comparison with the standard strategy of computing the Fitzsimmons ratio of the square root of the minimal positive Green’s function which is usually a hard task on graphs. Moreover, our method also results in weights which are larger at infinity than the weights computed with the standard method. 2. Manifolds Let us briefly comment on the strategy for obtaining optimal Hardy-type inequalities. Details will be provided in the next subsections. Take a function u>0 which is superharmonic, i.e., -\Delta u\geq 0 , and define the so-called Fitzsimmons ratio by V=-\Delta u/u . The existence of a positive superharmonic function is provided by the Agmon-Allegretto-Piepenbrink theorem. By standard theory, V is a Hardy weight. Since \sqrt{u} is also superharmonic, another Hardy weight is given by W:=-\Delta\sqrt{u}/\sqrt{u} . Additional restrictions on u then provide a certain optimality of W , cf. [ DFP14 , DP16 ] or Subsection 2.1 . These properties are naturally satisfied if u is just the (minimal positive) Green’s function. In this article, we restrict our considerations to spaces X with radially symmetric volume densities f . Consider u(r)=r/f(r),\qquad r=d(o,x),x\in S_{o}(r), where S_{o}(r) denotes the distance sphere with center o\in X and radius r>0 . The radial symmetry of the Laplacian applied to u leads to the superharmonicity of u . Hence, the corresponding function W is a Hardy weight. A deeper analysis also provides its optimality. Moreover, in certain cases, this ansatz leads to Poincaré-Hardy type inequalities with an explicit weight which is larger than the Fitzsimmons ratio of the Green’s function. Note that since V is generally not compactly supported, the fundamental theorem of optimality theory, cf. [ DFP14 , Theorem 2.2] or [ KPP20 , Theorem 9.2.1] , cannot be applied to obtain optimality. The fundamental theorem is actually cooked up for u being the Green’s function. However, this choice does not lead to Poincaré-Hardy-type inequalities on our manifolds, cf. [ BGG17 , FP25 ] for details. In the next subsection we recall basics of optimality theory. Thereafter, we show in detail the construction of optimal weights on \mathbb{H}^{d} . We close this section by giving extensions to models and harmonic manifolds. 2.1. A Short Reminder of Optimality Theory Let (X,g) be a smooth Riemannian manifold of dimension d and with measure \mu . Let -\Delta\geq 0 be the Laplace-Beltrami operator on (X,g) extended in the sense of Friedrichs to a self-adjoint operator on L^{2}(X,\mu) . Its quadratic form is defined via E(\varphi):=\int_{X}\left\lvert\nabla\varphi\right\rvert^{2}\,\mathrm{d}\mu,\qquad\varphi\in C^{\infty}_{c}(X). The correspondence becomes visible via Green’s formula, i.e., E(\varphi)=\left\langle-\Delta\varphi,\varphi\right\rangle,\qquad\varphi\in C^{\infty}_{c}(X). Here \left\langle\cdot,\cdot\right\rangle denotes the standard inner product on L^{2}(X,\mu) . Let \Omega\subseteq X be a domain and V a function. A function u\in C^{\infty}(\Omega) is called ( -\Delta+V )- (super-) harmonic if -\Delta u+Vu=0 (resp., -\Delta u+Vu\geq 0 ) in \Omega . A ( -\Delta )-(super-)harmonic function is simply called (super-)harmonic . Next, we recall the notion of optimality . It was stated first