
TL;DR
This paper addresses a specific question in symplectic topology related to Weinstein homotopies, providing partial insights into the structure and properties of Weinstein manifolds.
Contribution
It offers a partial answer to a question posed by Eliashberg concerning Weinstein homotopies, advancing understanding in symplectic topology.
Findings
Partial characterization of Weinstein homotopies
Insights into the structure of Weinstein manifolds
Progress towards answering Eliashberg's question
Abstract
We give a partial answer to a question asked by Eliashberg in one of his recent papers.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdvanced Topology and Set Theory · Advanced Topics in Algebra · Chronic Lymphocytic Leukemia Research
Weinstein Homotopies
Sauvik Mukherjee
Presidency University, Kolkata, India.
e-mail:[email protected]
Abstract.
In this paper we discuss a problem mentioned by Eliashberg in his paper [3]. He has asked if two completed Weinstein structures and on the same symplectic manifold can be homotoped through Weinstein structures. We discuss this problem and prove a weak partial result by assuming some additional conditions.
Key words and phrases:
Weinstein structures, Liouville homotopies
1. introduction
In this paper we discuss a problem mentioned by Eliashberg in his paper [3]. He has asked if two completed Weinstein structures and on the same symplectic manifold can be homotoped through Weinstein structures. We discuss this problem and prove a weak partial result by assuming some additional conditions.
We begin with the basic definitions. Let be a -dimensional symplectic domain with boundary with an exact symplectic form and primitive form i.e, .
A Liouville form is a choice of a primitive form such that is a contact form on and the orientation on by the form coincides with its orientation as the boundary of . The -dual vector field of is called the Liouville vector field. satisfies and hence its flow is conformally symplectically expanding.
Every Liouville domain can be completed in the following way.Set
[TABLE]
and extend on as on the attached end. Given a Liouville domain consider the compact set
[TABLE]
It is called Core or the Skeleton of the Liouville domain.
Let and be two Liouville forms on a fixed symplectic manifold , moreover let and be the respective Liouville vector fields. Then obviously for some and where is the hamiltonian vector field for .
A Liouville cobordism is a cobordism with an exact symplectic form such that the Liouville vector field points inward along and outward along .
Remark 1.1**.**
On the infinite end of , the Liouville vector field is given by irrespective of the choice of the Liouville form on .
Now we shall define the Weinstein structures. For this we need to recall few notions. A complete vector field is a vector field whose flow exists for all forward and backward time.
Let be a Morse function. A vector field is called gradient-like for if it satisfies
[TABLE]
for some and is with respect to some Riemannian metric and is with respect to its dual metric.
Definition 1.2**.**
*([7])
A Weinstein manifold is a symplectic manifold with a complete Liouville vector field which is gradient like with respect to the exhausting Morse function . A Weinstein cobordism is a Liouville cobordism whose Liouville vector field is gradient-like with respect to a Morse function which is constant on the boundary. A Weinstein cobordism with is called a Weinstein domain.*
In [3] Eliashberg has asked the following question.
Problem: Let and be two completed Weinstein structures on the same symplectic manifold . Are they homotopic as Weinstein structures ?
Obviously if and are the respective Liouville vector fields then for satisfying and hence
[TABLE]
gives a homotopy of Liouville vector fields. However may not be a Liouville homotopy. We refer the reader [7] for a precise definition of Liouville homotopy.
On Weinstein cobordisms a similar result has been proved in [7] although the Weinstein structures need to flexible. We refer the reader to [7] for a precise definition of flexible Weinstein structures.
Theorem 1.3**.**
*([7])
Let and be two flexible Weinstein structures on the same cobordism with dimension which coincide on . Let be a homotopy rel of non-degenerate two forms on connecting and . Then there exists a homotopy of flexible Weinstein structures connecting the given ones.*
Let us now return to the question asked by Eliashberg. We assume that all the zeros of are non-degenerate for all . So the zeros of executes curves (say). We consider and define vector field on . The curves ’s define curves ’s on as follows
[TABLE]
Consider two tubular neighborhoods of as . Let be cutoff functions such that and . Define on by canonically removing the zeros of as follows. Define close to as
[TABLE]
Let be the foliation defined by . Then is a regular foliation.
Definition 1.4**.**
We call the homotopy of the Liouville vector field uniformly open if it satisfies
- (1)
*All the zeros of are non-degenerate for all * 2. (2)
The foliation on is uniformly open
Please see 3.1 bellow for the definition of Uniformly open foliation. Now we state the main theorem of this paper.
Theorem 1.5**.**
Let and be two completed Weinstein structures on the same symplectic manifold and let the homotopy of the Liouville vector field is uniformly open (1.4), moreover and do not have a common zero. Then and can joined by a homotopy of Weinstein structures for which the underlying symplectic structure remains fixed.
Remark 1.6**.**
In 1.3 the underlying symplectic structure is not fixed.
2. -Principle
This section does not have any new result, we just recall some facts from the theory of -principle which we shall need in our proof.
Let be any fiber bundle and let be the space of -jets of jerms of sections of and be the -jet extension map of the section . If then is denoted as . A section is called holonomic if there exists a section such that . In the following we use the notation to denote a small open neighborhood of which is unspecified.
Let be a subset of . Then is called a differential relation of order . is said to satisfy -principle if any section can be homotopped to a holonomic section through sections whose images are contained in . Put differently, if the space of sections of landing into is denoted by and the space of holonomic sections of landing into is denoted by then satisfies -principle if the inclusion map induces a epimorphism at [math]-th homotopy group . satisfies parametric -principle if for all .
Let be a fiber bundle and by we denote the fiber preserving diffeomorphisms , i.e, if and only if there exists diffeomorphism such that the following diagram commutes
[TABLE]
Let be the projection . We call a fiber bundle natural if there exists a homomorphism such that . For a natural fiber bundle the associated jet bundle is also natural. The lift is given by
[TABLE]
where , , and is a local section near which represents the -jet . Observe and hence define .
For a natural fiber bundle , a differential relation is called -invariant if the action , leaves invariant.
Theorem 2.1**.**
*([4])
If a relation is open and -invariant on an open manifold then it satisfies parametric -principle.*
3. Bertelson’s Uniformly Open Foliations
In this section we recall some result from [1] and [2].
Definition 3.1**.**
*([1])
A foliated manifold is called uniformly open if there exists a function such that*
- (1)
* is proper,* 2. (2)
* has no leafwise local maxima,* 3. (3)
* is -generic.*
Remark 3.2**.**
Observe that if then can not be uniformly open as on a one dimensional manifold, a critical point will be either a local maximum or minimum.
So let us explain the notion -generic. In order to do so we need to define the singularity set for a map . is the set
[TABLE]
It was proved by Thom [6] that for most maps is a submanifold of . So we can restrict to and construct and so on. In [6] it has been proved that there exists such that .
Let us set as this is the only situation we need. Let be a foliated manifold with a leaf . Define the restriction map
[TABLE]
Define foliated analogue of the singularity set as
[TABLE]
Definition 3.3**.**
([1]) A smooth real valued function is called -generic if the first jet and the second jet for all .
Definition 3.4**.**
*([1])
An isotopy of the manifold is a family of diffeomorphisms of such that the map is smooth and . Consider a foliation on . A foliated isotopy of is an isotopy of that preserves the foliation , that is, for all . A relation is called foliated invariant on if the action by foliated isotopies leaves invariant.*
Theorem 3.5**.**
*([1])
On an uniformly open foliated manifold, any open, foliated invariant differential relation satisfies the parametric -principle.*
In [2] Bertelson has contructed counter examples that without the uniformly open condition 3.5 fails.
4. Main Theorem
In this section we prove 1.5. Let us first set some notations. First of all we have the Liouville vector fields and and let be the homotopy of uniformly open Liouville vector field. So we have
- (1)
2. (2)
With equality occurs in the above inequalities at the zeros of and . Define
[TABLE]
Then observe that
[TABLE]
Now consider
[TABLE]
So we get
[TABLE]
Recall that (according to 1.1) on the infinite end of the Liouville vector fields and are equal to . Moreover since and do not have a common zero and since ’s are compact is bounded bellow.
So the right hand side of the above inequality is bounded and hence the right hand side is equal to (say). So we get
[TABLE]
Without loss of generality we assume that otherwise we can use a relative version of -principle.
We replace by a new parameter where is such that on and on . We can replace the parameter in the above inequality.
Define one forms and as follows. First , , and . Similarly , and . So we have
[TABLE]
Now extending on and regularizing we get as in 1. We extend and to as . Adjusting and near ’s (1) to and so that
[TABLE]
Now we come to the -principle part. Consider and the trivial bundle where is the projection on the first factor. Observe that
[TABLE]
Note that this does not happen in case of higher order jet extensions as there will be mixed derivatives.
Observe that the section space . There is a natural affine fibration given by where . Define the relation as
[TABLE]
Obviously . Next we shall show that is open and invariant under -foliated isotopy. This will conclude the proof of 1.5 in view of 3.5. Only thing one needs to do is the following. Let is a resulting solution. Choose either or say . Then define as
[TABLE]
Now we have to re-introduce the singularities. Let be a family of Morse functions defined near with index same as the index of along . Let be a cutoff function such that on a tubular neighborhood and outside .Let . Observe
[TABLE]
Observe that has compact support and is of the form
[TABLE]
So if we take large enough then and obviously compactly supported. So we get the desired result.
Lemma 4.1**.**
The relation is open and invariant under the action of -foliated isotopies.
Proof.
Openness of follows directly from the definition of .
For second part we see , where is a positive real number. Positive as and is connected. So
[TABLE]
∎
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] Bertelson, Mélanie. A h ℎ h -principle for open relations invariant under foliated isotopies. J. Symplectic Geom. 1(2), 369425 (2002).
- 2[2] Bertelson, Mélanie Foliations associated to regular Poisson structures. Commun. Contemp. Math. 3 (2001), no. 3, 441–456. (Reviewer: Edith Padrón)
- 3[3] Eliashberg, Yakov. Weinstein manifolds revisited. ar Xiv.
- 4[4] Gromov, M. Partial Differential relations.
- 5[5] Eliashberg, Y.; Mishachev, N. Introduction to the h-principle. Graduate Studies in Mathematics, 48. American Mathematical Society, Providence, RI, 2002. xviii+206 pp. ISBN: 0-8218-3227-1 (Reviewer: John B. Etnyre)
- 6[6] R. Thom, Les singularités des application différentiables, Ann. Inst. Fourier (Grenoble), 6 (1955-1956) 43-87.
- 7[7] Cieliebak, Kai; Eliashberg, Yakov. From Stein to Weinstein and back , American Mathematical Society Colloquium Publications, vol. 59, American Mathematical Society, Providence, RI, 2012, Symplectic geometry of affine complex manifolds. MR 3012475 1,4,19,20,24,25
