Equivariant Morse-Bott cohomology through stabilization

Erkao Bao, Robi Huq, Shengzhen Ning · 2026

Read Equivariant Morse-Bott cohomology through stabilization by Erkao Bao, Robi Huq, Shengzhen Ning (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-2601-16119.txt.

Introduction

Erkao Bao, Robi Huq and Shengzhen Ning study a Morse view of group acts on shapes. Their 2026 preprint deals with shapes moved by a Lie group. A Morse tool reads shape traits from high and low points of a map. The team use a fix trick to make the map work with group acts. Small shifts of the map then give firm Morse Bott maps.

With firm maps, they build the group linked form of shape counts. They also gain the cross term rules the build needs. The build stems from the Austin and Braam path in the field. Firm maps close a gap in that path. The work adds rigour to group shape counts.

Catalog record

CREATORErkao Bao, Robi Huq, Shengzhen Ning
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE70 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-16119

Browse the full library · About the Tower of Babel

Full text

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

Equivariant Morse-Bott cohomology through stabilization Erkao Bao, Robi Huq, Shengzhen Ning Abstract . For closed manifolds with compact Lie group actions, we study Austin-Braam’s Morse-theoretic construction of Borel equivariant cohomology using the technique of stabilization . We show that a C 1 C^{1} -small equivariant perturbation produces stable invariant Morse-Bott functions. This allows us to realize the equivariant transversality and orientability assumptions in Austin–Braam’s framework by choosing generic invariant Riemannian metrics. 1. Introduction Equivariant Morse theory has been studied extensively since the pioneering work of Atiyah-Bott [ 1 ] on the Yang-Mills functional over Riemann surfaces. Equivariant Morse cohomology for general compact Lie group actions was first developed by Austin-Braam [ 2 ] . However, their construction relies on several assumptions (see Assumption 4.1 ) that do not hold in general. We now explain the main analytical difficulties underlying these assumptions. When the group G is not finite, G -invariant functions may have non-discrete critical sets and one must work with Morse-Bott functions instead of Morse functions. In the Morse-Bott setting, a typical chain model, used by [ 2 ] , associates to each critical submanifold its space of differential forms, and defines the boundary operator via pullback and integration along the fiber through the endpoint evaluation maps on the moduli spaces of gradient flow lines. From this perspective, the analytic foundation of the theory relies on establishing suitable transversality properties for relevant moduli spaces. The natural transversality condition in this context is the Morse-Bott-Smale condition, abbreviated MBS (Definition 1.7 ). There is also a weaker notion, referred to as weakly Morse-Bott-Smale , or weakly MBS (Definition 3.1 ). The first set of difficulties can be summarized as follows: (C1) While the weakly MBS condition can be achieved in the non-equivariant Morse-Bott setting, it generally fails in the equivariant Morse-Bott setting. Moreover, the weakly MBS condition alone is insufficient to define a Morse-theoretic chain complex. (C2) The MBS condition cannot, in general, be achieved even in the non-equivariant Morse-Bott case. A classical example in [ 8 ] , the torus with legs (Example 3.3 ), exhibits a Morse-Bott function for which the MBS condition fails for all metrics, although the weakly MBS condition can still be arranged. This example justifies (C2). Later, we introduce another example of a mapping torus (Example 5.2 ), which illustrates the obstruction described in (C1). A further complication in Morse-Bott theory is the following: (C3) The moduli spaces of gradient flow lines are not always orientable. This usually necessitates the use of local coefficient systems or orientation bundle valued differential forms. Taken together, these issues contribute to the relative underdevelopment of Morse-theoretic equivariant (co)homology for compact Lie group actions. We now analyze these difficulties in greater detail. The example of torus with legs (Example 3.3 ) can be upgraded into the equivariant setting by letting \mathbb{Z}_{n} act on it by rotations. It shows that the obstructions above persist even when the group action itself is well behaved. In particular, the \mathbb{Z}_{n} -action on torus with legs is free, yet the function still fails to admit a Morse-Bott-Smale metric. This observation motivates a refinement of the notion of G -invariant Morse-Bott functions. Rather than merely requiring G -invariance, we consider G -Morse-Bott functions (Definition 1.2 ), for which each critical submanifold is required to be a single G -orbit. Restricting attention to G -Morse-Bott functions excludes the example of \mathbb{Z}_{n} -equivariant torus with legs, but does not resolve the obstruction in (C1). Note that the function in the mapping torus example (Example 5.2 ) is G -Morse-Bott. To address this remaining difficulty, we adapt the stabilization approach developed in [ 9 , 5 ] . Starting from a G -Morse-Bott function, we perform a carefully chosen, C^{1} -small equivariant perturbation near the critical submanifolds. The resulting function, which we call stable , retains the essential equivariant features of the original function while exhibiting substantially improved analytical properties. It is worth emphasizing that, in the example of torus with legs, the Morse-Bott function is stable but not G -Morse-Bott. The combination of stability and the G -Morse-Bott condition turns out to be precisely what is needed. Key observation. The central theme of this paper is that stable G -Morse-Bott functions provide the appropriate framework of equivariant Morse theory for compact Lie group actions. By considering such class of functions, we resolve the transversality obstructions by achieving Morse-Bott-Smale condition for generic G -invariant metrics, and overcome orientability issues without appealing to local coefficient systems or orientation bundle valued differential forms. As a consequence, we obtain a well-defined Morse-theoretic equivariant cochain complex that recovers Borel equivariant cohomology. The stabilization procedure is explicit in certain cases, making concrete computations accessible. In a subsequent paper, we will exploit this feature further by studying the limiting behavior of the equivariant Morse complex as the size of the perturbation tends to zero, thereby relating the stabilized and unstabilized theories in a precise way. 1.1. Main results Let f\in C^{\infty}(M) be a smooth function on a closed smooth manifold M . Let p\in M be some critical point of f . The Hessian of f at the critical point p is a symmetric bilinear form Hess {}_{f}(p):T_{p}M\times T_{p}M\rightarrow\mathbb{R} defined by \text{Hess}_{f}(p)(v,w)=V(Wf), where V,W are any vector fields extending v,w\in T_{p}M . Let \text{ker}(\text{Hess}_{f}(p)):=\{v\in T_{p}M\,|\,\text{Hess}_{f}(p)(v,w)=0,\forall w\in T_{p}M\}. Definition 1.1 (Morse-Bott functions) . A smooth function f:M\to\mathbb{R} is called Morse-Bott if its critical set \crit(f) is a disjoint union of connected submanifolds such that for each connected component S of \crit(f) and any p\in S , we have T_{p}S=\text{ker}(\text{Hess}_{f}(p)) . We denote by \ind_{f}(S) (or simply \ind(S) when the function is clear in context) the Morse index of S , which is the maximal dimension of a subspace in T_{p}M where \text{Hess}_{f}(p) is negative definite, for any p\in S . Now, suppose there is a smooth action of the compact Lie group G on M . Let C^{\infty}(M)^{G} be the space of G -invariant smooth functions on M . When G is a Lie group, C^{\infty}(M)^{G} may not contain any Morse function since if p\in M is a critical point for some f\in C^{\infty}(M)^{G} , then all points in the orbit submanifold G\cdot p are also critical points of f . As a result, to study the G -equivariant Morse theory on M , it is natural to allow for critical submanifolds. Definition 1.2 ( G -Morse-Bott functions) . A G -invariant function f\in C^{\infty}(M)^{G} is called G -Morse-Bott if • f is Morse-Bott; • each connected component S of the critical submanifolds of f satisfies T_{p}S=T_{p}(G\cdot p) for any p\in S . This notion was simply called Morse in [ 13 ] or G -Morse in [ 3 ] . When G is a finite group, it also coincides with the notion of equivariant Morse function introduced in [ 5 , Definition 1.1] . If G acts trivially on M , then a G -Morse-Bott function reduces to an ordinary (non-equivariant) Morse function. It is important to distinguish between a G -invariant Morse-Bott function and a G -Morse-Bott function. For instance, constant functions are always G -invariant Morse-Bott, but they are not G -Morse-Bott in the sense of Definition 1.2 unless G acts transitively on

Read the complete text