ϕ \phi\, -SECTIONAL CURVATURE OF STATISTICAL STRUCTURES ON ALMOST CONTACT METRIC MANIFOLDS

Abbas Heydari, Sadegh Mohammadi · 2025

Read ϕ \phi\, -SECTIONAL CURVATURE OF STATISTICAL STRUCTURES ON ALMOST CONTACT METRIC MANIFOLDS by Abbas Heydari, Sadegh Mohammadi (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-2507-00811.txt.

Introduction

Abbas Heydari and Sadegh Mohammadi study a curvature measure on spaces that carry both a contact form and a statistical structure, in a paper from 2025. Such spaces arise in geometry where a metric is paired with twin connections. The measure gauges how sections aligned with the contact direction bend.

The authors prove this curvature is never positive on such spaces. They also give matching statements for when the curvature vanishes outright. From those, they derive a test that is enough to class the space as cosymplectic, a calm special case. The results add to the map of how statistics-flavored geometry behaves on contact spaces.

Catalog record

CREATORAbbas Heydari, Sadegh Mohammadi
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE37 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-00811

Browse the full library · About the Tower of Babel

Full text

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

ϕ \phi\, -SECTIONAL CURVATURE OF STATISTICAL STRUCTURES ON ALMOST CONTACT METRIC MANIFOLDS Abbas Heydari, Sadegh Mohammadi Abstract . In this article, we study and analyze the ϕ \phi -sectional curvature induced by a statistical structure on an almost contact metric manifold. We demonstrate that this sectional curvature is always non-positive. Additionally, we present equivalent statements regarding the vanishing of this type of sectional curvature. Furthermore, we derive a sufficient condition for an almost contact statistical manifold to be classified as a cosymplectic statistical manifold. 2020 Mathematics Subject Classification: 53B12, 53D15. Key words: statistical structures, sectional curvatures, almost contact metric manifolds, cosymplectic manifolds. 1. introduction Statistical structures were first introduced in the 1980s by S. L. Lauritzen in [ 13 ] . Later, in the article [ 12 ] , T. Kurose introduced an equivalent definition of statistical manifolds within the framework of affine differential geometry, which will be discussed in the following section. Statistical manifolds provide a geometric framework for studying and modeling probability distributions and play a significant role in various fields of science such as neutral networks, machine learning, information geometry, and physics [ 2 , 4 , 7 , 8 , 17 ] . For further reading in this area, Amari and Nagaoka’s textbook [ 3 ] and references [ 9 ] , [ 13 ] , and [ 6 ] are recommended. Statistical manifolds play a fundamental role in information geometry and the modeling of statistical data, while almost contact manifolds are widely used in Riemannian geometry to describe dynamical and physical systems. Combining these two structures creates a powerful framework that bridges statistical analysis with geometric modeling, offering new insights into complex systems where both statistical and physical components interact. This synthesis not only enhances our understanding of such systems but also opens new avenues for studying nonlinear dynamics and predictive modeling. The concept of the almost contact statistical manifolds was introduced by F. Malek in the article [ 1 ] . One of the main approaches to studying and identifying manifolds is through their curvature. Since sectional curvature plays a crucial role in understanding a manifold’s geometric structure, its study is of great importance. Sectional curvature is a fundamental concept in differential geometry, primarily associated with Riemannian metrics. While the curvature tensor can be defined for any connection, sectional curvature— which assigns a numerical value to a tangent space plane—typically requires a metric tensor. A key observation is that if a tensor can be defined in relation to the Riemannian metric in the same way as the curvature tensor, then the concept of sectional curvature can be extended. This idea was successfully explored in the work of B. Opozda, who introduced statistical sectional curvature corresponding to any statistical structure on a manifold [ 16 ] . In this paper, we investigate the sectional curvature associated with an almost contact statistical structure on a manifold and establish that this curvature is always non-positive. Additionally, we connect this concept to previous research on statistical structures in almost contact metric manifolds, emphasizing its relevance in understanding their geometric properties. Specifically, we show that the vanishing of this sectional curvature reveals essential characteristics of the statistical structure on an almost contact metric manifold. Finally, we demonstrate that if the statistical connection is compatible with the almost contact metric structure, then every almost contact statistical manifold must be a cosymplectic statistical manifold. 2. Statistical Manifolds Statistical structures on a manifold M can be defined through three equivalent approaches. Firstly, a statistical manifold is a Riemannian manifold (M,g) equipped with a symmetric 3-covariant tensor field [ 13 ] . For an alternative equivalent definition of statistical manifolds, consider the pair (g,\nabla) , where g\in\otimes^{0}_{2}M is a Riemannian metric tensor on M , and \nabla is a torsion-free affine connection such that the associated tensor field \nabla g\in\otimes^{0}_{3}M is symmetric. In this case, the pair (g,\nabla) is called a statistical structure on M , and the triplet (M,g,\nabla) is known as a statistical manifold [ 12 ] . Every Riemannian manifold (M,g) , along with the associated Levi-Civita connection, is a typical example of statistical manifolds. In other words, statistical manifolds can be considered as an extension of Riemannian manifolds. An alternative but equivalent formulation of statistical manifolds states that the triplet (M,g,K) defines a statistical manifold, where g\in\otimes^{0}_{2}M is a Riemannian metric tensor and K\in S^{2}(M;TM) is a symmetric tensor field on M that is symmetric with respect to g . The symmetry of the tensor field K with respect to the Riemannian metric tensor g means that the cubic form C(X,Y,Z):=g(X,K(Y,Z)) is symmetric. The validity of this equivalence is established by jointly considering the mathematical fact that the set of all torsion-free affine connections on the manifold M forms an affine space whose associated vector space is S^{2}(M;TM) , along with the existence of musical isomorphisms in the Riemannian manifold (M,g) . Furthermore, the statistical connection \nabla , the Levi-Civita connection \nabla^{\circ} , the tensor field K , and the Riemannian metric g satisfy the relationships K=\nabla-\nabla^{\circ} and g(X,K(Y,Z))=-2(\nabla_{X}g)(Y,Z) . 3. Almost Contact Metric Manifolds This section is based on [ 5 ] and has been written accordingly. Definition 3.1 . Let M be an odd-dimensional manifold. The triplet (\phi,\xi,\eta) , where \phi is a tensor field of type (1,1), \xi is a vector field, and \eta is a 1-form on M , is called an almost contact structure on M , provided that \phi^{2}=-I+\eta\otimes\xi\quad,\quad\eta(\xi)=1 are satisfied. If (\phi,\xi,\eta) is an almost contact structure on the manifold M^{2n+1} , then \displaystyle\phi(\xi)=0\quad,\quad\eta\circ\phi=0\quad,\quad rank(\phi)=2n hold true. Definition 3.2 . Suppose (\phi,\xi,\eta) is an almost contact structure on the manifold M^{2n+1} . If there exists a Riemannian metric g\in\otimes^{0}_{2}M such that \displaystyle\forall X,Y\in\mathfrak{X}(M);g(\phi X,\phi Y)=g(X,Y)-\eta(X)\eta(Y) holds, then the quadruple (\phi,\xi,\eta,g) defines an almost contact metric manifold. It is straightforward to see that if (M,\phi,\xi,\eta,g) is an almost contact metric manifold, then \displaystyle g(\xi,\xi)=1\quad,\quad\eta(X)=g(X,\xi)\quad,\quad g(\phi X,Y)+g(X,\phi Y)=0. Definition 3.3 . An almost contact metric manifold (M,\phi,\xi,\eta,g) is called a cosymplectic manifold if \nabla^{\circ}\phi=0 , where \nabla^{\circ} is the unique Levi-Civita connection associated with the Riemannian manifold (M,g) . 4. Almost Contact Statistical Manifolds Definition 4.1 . Suppose (M,\phi,\xi,\eta,g) is an almost contact metric manifold, and (g,\nabla=\nabla^{\circ}+K) is a statistical structure on M . Then, (M,\phi,\xi,\eta,g,\nabla=\nabla^{\circ}+K) is called an almost contact statistical manifold if \displaystyle\forall X,Y\in\mathfrak{X}(M);K(X,\phi Y)+\phi K(X,Y)=0 is satisfied [ 1 ] . Proposition 4.2 . If (M,\phi,\xi,\eta,g) is an almost contact metric manifold and (g,\nabla=\nabla^{\circ}+K) is a statistical structure on M , then (M,\phi,\xi,\eta,g,\nabla=\nabla^{\circ}+K) is an almost contact statistical manifold if, and only if, \displaystyle\forall X,Y\in\mathfrak{X}(M);K(X,\phi Y)=K(\phi X,Y). Proof. If (M,\phi,\xi,\eta,g,\nabla=\nabla^{\circ}+K) is an almost contact statistical manifold, then \displaystyle K(X,\phi Y)=-\phi K(Y,X)=K(\phi X,Y) holds. Conversely, if the statistical structure (g,\nabla=\nabla^{\circ}+K) on the almost contact metric manifold (M,\phi,\xi,\eta,g) is such that K(X,\phi Y)=K(\phi X,Y) is satisfied, then \displaystyle g(K(\phi X,Y),Z) \displaystyle=g(K(\phi X,Z),Y)=g(K(X,\phi Z),Y) \displaystyle=g(K(X,Y),\phi Z)=g(-\phi K(X,Y),Z). Now, based on the nondegeneracy of the Riemannian metric tensor g , we conclude that K(\phi X,Y)=-\phi K(X,Y) holds. Thus, the statistical structure (g,\nabla=\nabla^{\circ}+K) on (M,\phi,\xi,\eta,g) defines an almost contact statistical structure. \blacksquare Lemma 4.3 . [ 14 ] Assume that (M,\phi,\xi,\eta,g,\nabla=\nabla^{\circ}+K) is an almost contact statistical manifold. Therefore, we have K(X,\xi)=\lambda\eta(X)\xi , where \lambda=g(K(\xi,\xi),\xi)\in C^{\infty}(M) .

Read the complete text