Alexander-Taylor’s inequality for capacities in complex Sobolev spaces

Ngoc Cuong Nguyen, Do Duc Thai · 2026

Read Alexander-Taylor’s inequality for capacities in complex Sobolev spaces by Ngoc Cuong Nguyen, Do Duc Thai (2026) 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-2603-05838.txt.

Introduction

Nguyen and Thai prove a sharp match-up of Alexander-Taylor size against working size, working in a complex Sobolev space over a compact Kähler shape. Dinh, Sibony, and Vigny gave the latter space and size. Their result gives a precise match-up between the two sizes in this shape setting, tightening the link between them.

Catalog record

CREATORNgoc Cuong Nguyen, Do Duc Thai
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE40084
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.05838

Browse the full library · About the Tower of Babel

Full text

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

Alexander-Taylor’s inequality for capacities in complex Sobolev spaces Ngoc Cuong Nguyen, Do Duc Thai Abstract . We prove a sharp inequality between the Alexander-Taylor capacity and the functional capacity in a complex Sobolev space on a compact Kähler manifold. The latter space and capacity were introduced by Dinh, Sibony and Vigny. 1. Introduction In 1980s Alexander and Taylor [ AT84 ] proved a sharp quantitative comparison between the relative capacity of Bedford and Taylor [ BT82 ] and the projective capacity of Alexander [ A79 ] . It is now called the Alexander-Taylor inequality in the literature, which has become very useful. For example, it gave in [ AT84 ] a simplified proof of an effective version of Josefson’s theorem, due to El Mir [ E79 ] , about the equivalence between locally pluripolar sets and globally pluripolar sets in \mathbb{C}^{n} . Subsequently, the inequality provided a crucial link in proving almost sharp uniform estimates for complex Monge-Ampère equations on both domains in \mathbb{C}^{n} and compact Kähler manifolds via the capacity method [ K98 , K05 ] . Its generalization to the equations with semi-positive background forms on compact Kähler manifolds played equally important roles in [ GZ05 , GZ17 ] and [ EGZ09 ] . Let (X,\alpha) be a compact Kähler manifold of dimension n . Let \omega\geq 0 be a semi-positive closed (1,1) -form such that \int_{X}\omega^{n}>0 . For simplicity it is normalized by \int_{X}\omega^{n}=1 . Denote by PSH(X,\omega) the family of \omega -plurisubharmonic ( \omega -psh) functions on X . The global Alexander-Taylor capacity T_{\omega}(\cdot) introduced and studied by Guedj and Zeriahi [ GZ05 ] (see also [ DS06b ] and [ GZ17 ] ) is as follows. For a compact subset K\subset X , T_{\omega}(K)=\exp\left(-\sup_{X}V_{K}^{*}\right), where V_{K}(z)=\sup\{v(z):v\in PSH(X,\omega),v\leq 0\text{ on }K\} and V_{K}^{*} is its upper semi-continuous regularization. The latter function is often called the Siciak-Zaharjuta extremal function associated to K . We are interested in the subspace W^{*}(X)\subset W^{1,2}(X,\mathbb{R}) introduced by Dinh and Sibony [ DS06a ] (see Section 2), which is called the complex Sobolev space. For f\in W^{*}(X) , let \Gamma_{f} denote the non-empty set of all positive closed (1,1) -currents T satisfying df\wedge d^{c}f\leq T\quad\text{weakly in }X. Then, the W^{*} -norm of f is given by \|f\|_{*}^{2}=\|f\|_{L^{2}(X)}^{2}+\inf\{\|T\|:T\in\Gamma_{f}\}, where \|T\|=\int_{X}T\wedge\alpha^{n-1}. Soon afterward, Vigny [ Vig07 ] showed that W^{*}(X) is a Banach space with the W^{*} -norm. One special feature of this new space is that it takes complex structures of underlying manifolds into account. Therefore, it is a suitable framework to work with complex objects such as holomorphic/meromorphic maps and quasi plurisubharmonic functions. Since then, this space has found numerous applications in complex dynamics in higher dimensions and some related areas [ BiD23 ] , [ DKW21 ] , [ DoNV25 ] , [ Vig17 ] , [ Vu20 ] and [ WZ24 ] . Since W^{*}(X) is a Banach space, it has a natural functional capacity {\tt c}(\cdot) defined by its norm, which was introduced in [ Vig07 ] . For a Borel subset E\subset X , {\tt c}(E)=\inf\left\{\|v\|_{*}^{2}\;\rvert\;v\in\mathscr{K}(E)\right\}, where \mathscr{K}(E)=\left\{v\in W^{*}(X)\;\rvert\;\{v\leq-1\}^{o}\supset E\text{ and }v\leq 0\right\}. The inclusion \{v\leq-1\}^{o}\supset E means v\leq-1 a.e. in a neighborhood of E . In fact, this is a capacity in Choquet’s sense [ Vig07 , Theorem 30] . It should be pointed out that the functional capacity of with respect to the Sobolev norm is not strong enough to study properties of functions in W^{*}(X) . The reason is that there are bounded and unbounded \omega -psh functions belong to W^{*}(X) while the Sobolev capacity cannot be used to characterize neither quasicontinuity of such functions nor pluripolar sets. The main result of the paper is the following comparison between the global Alexander-Taylor capacity and the functional capacity of W^{*}(X) above. Theorem 1.1 . There exists a constant A>0 such that for every compact subset K\subset X , \exp\left(-A/{\tt c}(K)\right)\leq T_{\omega}(K)\leq\exp\left(-1/(4n)^{\frac{1}{n}}[{\tt c}(K)]^{\frac{1}{n}}\right). Since both {\tt c}(\cdot) and T_{\omega}(\cdot) are Choquet capacities, the inequalities hold for all Borel sets. Especially, they keep the spirit of the ones in [ GZ17 , Lemmas 12.2, 12.3] where T_{\omega}(K) is compared with the global Bedford-Taylor capacity cap_{\omega}(K) . Namely, in the first inequality the constant A is equivalent to the one of [ GZ17 , Lemma 12.3] , up to a numerical constant. In the second one, it is significant that the constant on the right hand side is purely a dimensional constant which is similar to ones in the local setting [ AT84 ] and global setting [ GZ17 , Lemma 12.1] . Regarding to the exponents in the inequalities, Remark 3.5 shows that they are sharp. We expect that this inequality will be a bridge so that the tools from complex Sobolev spaces can be used to study open problems in pluripotential theory and complex Monge-Ampère equations. Some evidences have been appeared very recently in [ DoN25 ] , [ DKN22 ] , [ VV24 ] and [ Vu24 , Vu26 ] . We give here one more application to the equation with a background form \omega which is semi-positive and big . This result was implicitly contained in [ DKN22 , Theorem 3.6] . Corollary 1.2 . Let \mu be a probability measure on X . Assume \mu is (W^{*}(X),\log^{p}) -continuous with p>n+1 , i.e., there is constants c>0 such that \left|\int_{X}fd\mu\right|\leq c\left[1+|\log\|f\|_{L^{1}(X)}|\right]^{-p} for every f\in W^{*}(X)\cap C^{0}(X) and \|f\|_{*}\leq 1 . Then, there exists a bounded \omega -psh solution u to (\omega+dd^{c}u)^{n}=\mu. Organization. In Section 2 we recall basic facts of pluripotential theory and complex Sobolev spaces. In particular, we give another proof of existence of quasi-continuous representative for bounded functions in W^{*}(X) . Section 3 devotes to prove the main theorem and corollary. Acknowledgement. A part of the paper was completed while the first author was visiting University of Engineering and Technology (UET) in Hanoi in December 2025. He wish to thank Pham Hoang Hiep for a warm invitation. The paper was written while the first author visited the Center for Complex Geometry (Daejeon). He would like to thank Jun-Muk Hwang and Yongnam Lee for their kind support and exceptional hospitality. He is also grateful to the institution for providing perfect working conditions. The research of the second author is supported by the ministry-level project B2025-SPH-02. 2. A functional capacity in complex Sobolev spaces In this section we recall some basic facts about the complex Sobolev space and pluripotential theory. Let (X,\alpha) be a compact Kähler manifold of dimension n . Let \omega\geq 0 be a semi-positive closed (1,1) -form whose volume is normalized by (2.1) \int_{X}\omega^{n}=1. Clearly, we may assume that (2.2) \omega\leq\alpha. Let W^{1,2}(X,\mathbb{R}) denote the Sobolev space of functions in L^{2}(X,\mathbb{R}) whose first order (weak) derivatives are L^{2} -integrable. For f\in W^{1,2}(X,\mathbb{R}), let \Gamma_{f} denote the set of all positive closed (1,1) -currents T satisfying (2.3) df\wedge d^{c}f\leq T\quad\text{weakly in }X. Here we use the normalization d^{c}=\frac{\sqrt{-1}}{2\pi}(\overline{\partial}-\partial),\quad dd^{c}=\frac{\sqrt{-1}}{\pi}\partial\overline{\partial}. Definition 2.1 ( [ DS06a ] ) . A function f\in W^{1,2}(X,\mathbb{R}) belongs to the complex Sobolev space W^{*}(X) if the set \Gamma_{f} associated to f is non-empty. The total mass of a closed positive (1,1)-current T on X is given by \|T\|=\int_{X}T\wedge\alpha^{n-1}. By the compactness of X and closedness of T , \|T\|

Read the complete text