TL;DR
This paper extends the classical Orlik-Solomon theory for hyperplane arrangements to the more general setting of toric arrangements. The authors provide explicit presentations for the cohomology algebra of the complement of any toric arrangement, generalizing previous results for the unimodular case and proving integral formality.
What Problem Does This Paper Address?
The study of the topology of complements of arrangements of linear subspaces (hyperplanes) in vector spaces has a rich history, with the Orlik-Solomon algebra providing a powerful combinatorial description of their cohomology. This theory has been successfully generalized to "toric arrangements" – arrangements of subtori in a complex torus. However, existing work primarily focused on "unimodular" toric arrangements or provided results only for rational coefficients. This paper tackles the challenge of providing a unified, explicit presentation for the integral cohomology algebra of any toric arrangement, extending the Orlik-Solomon framework beyond the unimodular case and rational coefficients. A key difficulty is that the standard combinatorial data (matroids) for hyperplane arrangements is insufficient for toric arrangements, requiring a richer set of combinatorial invariants.
Methods
The authors build upon existing work by De Concini and Procesi, as well as Dupont. They introduce and utilize "logarithmic forms" associated with the characters defining the toric arrangement. A crucial technique involves constructing specific "separating coverings" of the torus. These coverings allow them to relate the cohomology of the general toric arrangement to that of unimodular ones, enabling the extension of known results. They leverage formal identities among these logarithmic forms and carefully analyze the combinatorial data encoded in the poset of connected components of intersections. The presentation is constructed using generators and relations, mirroring the structure of the Orlik-Solomon algebra.
Datasets
Not discussed in the paper. The paper deals with theoretical constructions and combinatorial objects rather than empirical datasets.
Key Insights and Ideas
- Generalization of Orlik-Solomon Theory: The paper successfully extends the powerful Orlik-Solomon framework to the broader class of toric arrangements, providing a concrete way to understand their cohomology.
- Integral Formality: A significant result is the extension of Dupont's rational formality result to integral formality, meaning the cohomology algebra is isomorphic to a differential graded algebra whose generators and relations are explicitly given.
- Combinatorial Encoding: The authors demonstrate that the data needed for the rational cohomology presentation is fully encoded in the poset of connected components of intersections of the arrangement. This highlights the deep connection between the combinatorial structure and the topological invariants.
- Beyond Unimodularity: The work explicitly addresses the challenges posed by non-unimodular toric arrangements, where the standard combinatorial tools are insufficient, requiring more sophisticated techniques like separating coverings.
- Role of Logarithmic Forms: The paper showcases the utility of carefully chosen logarithmic forms as generators for the cohomology algebra, providing a concrete algebraic structure.
Key Results
- An explicit Orlik-Solomon type presentation for the integral cohomology algebra of any toric arrangement is provided.
- This presentation generalizes the work of De Concini and Procesi for unimodular toric arrangements.
- The integral cohomology algebra is shown to be formally isomorphic to a differential graded algebra.
- The authors prove that the forms used as generators are integral, and the relations hold as formal relations between these forms, not just as relations between cohomology classes.
Limitations and Future Work
- Data Dependency: The paper emphasizes that while the rational cohomology presentation is fully determined by the poset of connected components of intersections, the integral cohomology is not solely determined by this poset. This suggests that further combinatorial invariants might be needed for a complete understanding of the integral case.
- Global Coverings: The authors note that constructing a global unimodular covering for any toric arrangement is not always possible, indicating a limitation in directly applying unimodular techniques in all cases.
- Future Directions: The paper's results open avenues for further research into the interplay between combinatorics and topology in toric arrangements, particularly in understanding the nuances of integral cohomology and exploring connections to other areas like representation theory and algebraic geometry. The authors also point to the possibility of extending these results to other types of arrangements.
Summary updated