Abstract
This paper provides a complete classification of the singularity categories for cluster-tilted algebras of Dynkin types $A$, $D$, and $E$. By leveraging the stable categories of self-injective Nakayama algebras, the authors demonstrate that these complex homological structures can be decomposed into well-understood building blocks. This work effectively settles the problem of singularity equivalence classification for these fundamental algebraic structures.
Introduction: Measuring Algebraic "Roughness"
In the field of representation theory, a central goal is to understand the homological properties of algebras. A "smooth" algebra is one with finite global dimension, meaning every module can be resolved by a finite sequence of projective modules. However, many important algebras—including cluster-tilted algebras—possess infinite global dimension. To study the "rough" parts of these algebras that remain after accounting for projectives, mathematicians use the singularity category (introduced by Buchweitz and Orlov).
Cluster-tilted algebras are a significant class of algebras arising from cluster categories. While their derived equivalence classification (how they behave in derived categories) was largely understood, their singularity equivalence classification remained an open problem. This paper addresses this gap for algebras associated with the classic Dynkin diagrams ($A_n, D_n, E_6, E_7, E_8$), which are foundational to Lie theory and geometry.
Methodology: Mutations and Gluing
The authors employ several sophisticated tools from homological algebra to tackle the classification problem. A key property utilized is that cluster-tilted algebras are 1-Gorenstein. For such algebras, the singularity category is equivalent to the stable category of Gorenstein projective modules, providing a more concrete way to perform calculations.
One of the primary techniques used is the "gluing" of quivers. By representing complex quivers as combinations of simpler rooted quivers, the authors can analyze how the singularity category changes as these components are joined.
In addition to gluing, the authors use "good mutations"—algebraic operations that transform the quiver of an algebra while preserving its derived category. This allows them to reduce a wide variety of complex algebras into a few "standard forms" that are easier to classify.
Key Insights: The Role of Triangles and Nakayama Algebras
The most significant insight of the paper is that the singularity categories of these complex algebras often decompose into a direct product of very simple categories. Specifically, they identify the stable categories of self-injective Nakayama algebras ($S_l$) as the fundamental units of these singularities.
Type A and the Geometry of Triangles
For cluster-tilted algebras of type $A$, the "amount" of singularity is directly proportional to the number of triangles (3-cycles) present in the algebra's quiver. Each triangle contributes exactly one copy of the stable category of the Nakayama algebra $S_3$ to the total singularity category.
This geometric intuition allows for a remarkably simple classification: the singularity category of a type $A$ algebra is equivalent to $\prod_{t(Q)} \underline{\text{mod}}(S_3)$, where $t(Q)$ is the number of triangles.
Numerical Invariants for Type E
For the exceptional types ($E_6, E_7, E_8$), the authors discovered that the singularity category is uniquely determined by two numerical "fingerprints":
- The determinant of the Cartan matrix.
- The trace of the asymmetry matrix.
They prove that two such algebras are singularity equivalent if and only if these two values match, providing a powerful numerical shortcut for classifying highly complex categories.
Summary of Results
The paper successfully classifies the singularity categories for all three Dynkin types:
- Type A: The category is a product of $\underline{\text{mod}}(S_3)$ based on the triangle count.
- Type D: The authors identify four distinct families (Types I-IV). Their singularity categories are products of $\underline{\text{mod}}(S_3)$ and other specific Nakayama algebras like $\underline{\text{mod}}(S_m)$.
- Type E: A complete mapping is provided for $E_6, E_7,$ and $E_8$, linking specific associated polynomials to their corresponding singularity categories via the Cartan and asymmetry invariants.
Conclusion and Future Directions
By reducing the problem of singularity equivalence to the study of Nakayama algebras and numerical invariants, this paper provides a definitive map of the homological "roughness" of cluster-tilted algebras of finite type. While this settles the Dynkin case, future research may extend these methods to Euclidean types (tame representation type) or more general cluster-tilted algebras of infinite type, potentially uncovering deeper geometric connections between these categories and cluster varieties.
Summary updated