Combinatorial Heegaard Floer homology and nice Heegaard diagrams (Q436168): Difference between revisions

From the authors' introduction: Heegaard Floer homology is an invariant for 3-manifolds, defined by Ozsváth and Szabó using a Heegaard diagram for the 3-manifold. Its definition rests on a suitable adaptation of Lagrangian Floer homology in a symmetric product of the Heegaard surface, relative to embedded tori which are associated to the attaching circles. These Floer homology groups have several versions. The simplest version \( \widehat {\text{HF}} (Y)\) is a finitely generated Abelian group, while \(HF^{-} (Y)\) admits the algebraic structure of a finitely generated \(Z [U]\)-module. Building on these constructions, one can define invariants of knots and links in \(3\)-manifolds, invariants of smooth \(4\)-manifolds, contact structures, sutured \(3\)-manifolds, and 3-manifolds with parameterized boundary. The invariants are computed as homology groups of certain chain complexes. The definition of these chain complexes uses a choice of a Heegaard diagram of the given 3-manifold, and various further choices (e.g., an almost complex structure on the symmetric power of the Heegaard surface). Both the definition of the boundary map and the proof of independence of the homology from these choices involve analytic methods. Sarkar and Wang discovered that by choosing an appropriate class of Heegaard diagrams for \(Y\) (which they called \textit{nice}), the chain complex computing the simplest version \( \widehat {\text{HF}} (Y)\) can be explicitly computed. In addition, Sarkar and Wang also showed that every closed 3-manifold admits a nice Heegaard diagram. In a similar spirit, it was shown that all versions of the link Floer homology groups for links in \(S^3\) admit combinatorial descriptions using grid diagrams. Indeed the topological invariance of this combinatorial description of link Floer homology is verified using direct combinatorial methods (and, in particular, avoiding analysis). The aim of the present work is to develop a stabilized version of Heegaard Floer homology which uses only combinatorial/topological methods, and in particular is independent of the theory of pseudo-holomorphic disks. As part of this, we construct a class of Heegaard diagrams for closed, oriented 3-manifolds which are naturally associated to pair-of-pants decompositions. The bulk of this paper is devoted to a direct, topological proof of the topological invariance of the resulting Heegaard Floer invariants. The three primary objectives of this paper are the following: 1) to give an effective construction of Heegaard diagrams for 3-manifolds for which a chain complex computing the stabilized version of Heegaard Floer homology can be explicitly described; 2) to give some relationships between Heegaard Floer homology with more classical objects in 3-manifold topology (specifically, pair-of-pants decompositions for Heegaard splittings). The authors hope that further investigations along these lines may shed light on topological properties of Heegaard Floer homology; 3) to give a self-contained, topological description of some version of Heegaard Floer homology. One might hope that the outlines of this approach could be applied to studying other Floer-homological 3-manifold invariants.