Categorical properties and homological conjectures for bounded extensions of algebras

Yongyun Qin, Xiaoxiao Xu, Jinbi Zhang, Guodong ZhouarXiv 2024 · DOI · arXiv

Abstract

An extension $B\subset A$ of finite dimensional algebras is bounded if the $B$-$B$-bimodule $A/B$ is $B$-tensor nilpotent, its projective dimension is finite and $\mathrm{Tor}_i^B(A/B, (A/B)^{\otimes_B j})=0$ for all $i, j\geq 1$. We show that for a bounded extension $B\subset A$, the algebras $A$ and $B$ are singularly equivalent of Morita type with level. Additionally, under mild conditions, their stable categories of Gorenstein projective modules and Gorenstein defect categories are equivalent, respectively. Some homological conjectures are also investigated for bounded extensions, including Auslander-Reiten conjecture, finististic dimension conjecture, Fg condition, Han's conjecture, and Keller's conjecture. Applications to trivial extensions and triangular matrix algebras are given.

Introduction: The Problem of Algebraic Singularity

In the study of finite-dimensional algebras, a central goal is to understand the "singularity" of an algebra—a measure of how far it deviates from being "smooth" (having finite global dimension). This is formally captured by the singularity category. A persistent challenge in homological algebra is determining when two distinct algebras share the same homological "DNA."

This paper investigates algebra extensions $B \subset A$. The core question is: if we understand the properties of a smaller algebra $B$, what can we definitively say about the larger algebra $A$? The authors introduce and explore bounded extensions, a class of extensions designed to preserve complex categorical structures and homological invariants.

Methodology: Bridging Categories

The research employs advanced tools from homological algebra and derived category theory:

  • Singular Equivalence of Morita Type with Level: This generalizes classical stable equivalences. By using bimodules to bridge categories and incorporating a "level" $l$ related to the syzygy functor, the authors can handle extensions where bimodules are not strictly projective.
  • Verdier Quotients: These are used to define singularity and Gorenstein defect categories by "factoring out" projective modules.
  • Adjoint Functors: The authors construct sequences of adjoint functors between the derived categories of $A$ and $B$ to establish equivalences between their respective singularity categories.
  • Handy Criteria: A significant contribution is the development of Criterion 0.4, which provides conditions for a functor between module categories to induce a triangle functor between stable categories of Gorenstein projective modules.

Key Insights: The Three Pillars of Boundedness

The authors identify three specific conditions that define a "bounded" extension $B \subset A$:

  1. The quotient $A/B$ must be tensor nilpotent as a $B$-$B$-bimodule.
  2. The projective dimension of $A/B$ as a bimodule must be finite.
  3. A vanishing Tor condition must be satisfied: $\mathrm{Tor}_i^B(A/B, (A/B)^{\otimes_B j})=0$ for all $i, j \geq 1$.

These conditions ensure that the homological fates of $A$ and $B$ are linked. Furthermore, by utilizing singular equivalences "with level," the authors can accommodate extensions that were previously difficult to analyze, using the "level" as a buffer for non-projective bimodules.

Key Results and Conjectures

The paper establishes several landmark results:

  • Theorem 0.3: Proves that a bounded extension $B \subset A$ induces a singular equivalence of Morita type with level, meaning $A$ and $B$ are homologically identical in their singular parts.
  • Invariance of Conjectures (Theorem 0.5): The authors demonstrate that several major open problems are invariant under bounded extensions. Specifically, if $B \subset A$ is bounded, then the Finitistic Dimension Conjecture, Han’s Conjecture, and Keller’s Conjecture hold for $B$ if and only if they hold for $A$.
  • Gorenstein Symmetry: Under mild conditions (e.g., $B$ being a direct summand of $A$), the Gorenstein defect category—which measures the distance from being a Gorenstein algebra—is equivalent for both $A$ and $B$.

Practical Applications

The theoretical results are applied to common algebraic structures, such as triangular matrix algebras and trivial extensions.

Figure 1: Quiver representation of a triangular matrix algebra, a primary application for bounded extensions.
Figure 2: Quiver with a cycle, representing more complex structures like trivial extensions that can be analyzed as bounded extensions.

These applications show that bounded extensions provide a reliable reduction method: to prove a conjecture for a complex algebra $A$, one can often simplify the problem by proving it for a sub-algebra $B$, provided the extension is bounded.

Limitations and Future Work

While the singularity category equivalence is broad, equivalences for Gorenstein projective modules require stricter assumptions. The authors also highlight that not all extensions are bounded, citing relation extensions of tilted algebras as examples that fall outside this framework. Future work is expected to apply these "handy criteria" to other categorical structures like recollements and Morita context algebras.

Summary updated

Abstract

An extension $B\subset A$ of finite dimensional algebras is bounded if the $B$-$B$-bimodule $A/B$ is $B$-tensor nilpotent, its projective dimension is finite and $\mathrm{Tor}_i^B(A/B, (A/B)^{\otimes_B j})=0$ for all $i, j\geq 1$. We show that for a bounded extension $B\subset A$, the algebras $A$ and $B$ are singularly equivalent of Morita type with level. Additionally, under mild conditions, their stable categories of Gorenstein projective modules and Gorenstein defect categories are equivalent, respectively. Some homological conjectures are also investigated for bounded extensions, including Auslander-Reiten conjecture, finististic dimension conjecture, Fg condition, Han's conjecture, and Keller's conjecture. Applications to trivial extensions and triangular matrix algebras are given. In course of proof, we give some handy criteria for a functor between module categories to induce triangle functors between stable categories of Gorenstein projective modules and Gorenstein defect categories, which generalise some known criteria, and hence might be of independent interest.
BibTeX
@misc{qin2024categorical,
  author = {Yongyun Qin and Xiaoxiao Xu and Jinbi Zhang and Guodong Zhou},
  title = {{Categorical properties and homological conjectures for bounded extensions of algebras}},
  year = {2024},
  eprint = {2407.21480},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2407.21480},
  howpublished = {arXiv},
}

Loading related content...

Figure