Exceptional sequences and rooted labeled forests

Basic Information

• Paper ID: 2108.11351
• Title: Exceptional sequences and rooted labeled forests
• Authors: Kiyoshi Igusa, Emre Sen
• Classification: math.RT (Representation Theory), math.CO (Combinatorics)
• Publication Date: arXiv preprint, latest version December 31, 2024
• Paper Link: https://arxiv.org/abs/2108.11351

Abstract

This paper establishes a representation-theoretic bijection connecting rooted labeled forests with $n$ vertices to complete exceptional sequences of the linearly oriented quiver of type $A_n$. Ascending and descending vertices in the forest correspond respectively to relatively injective and relatively projective objects in the exceptional sequence. The authors prove that each object in an exceptional sequence of the linearly oriented $A_n$ is either relatively projective or relatively injective (or both). The paper constructs a natural action of the extended braid group on rooted labeled forests and proves its consistency with the known action of the braid group on complete exceptional sequences. Additionally, it describes the action of the Garside element $\Delta$ of the braid group on rooted labeled forests and demonstrates its relationship with cluster theory.

Research Background and Motivation

Problem Background

1. Combinatorial interpretation of exceptional sequences: For Dynkin quivers of type $A_n$, there exist $(n+1)^{n-1}$ complete exceptional sequences. From a combinatorial perspective, this count appears in numerous enumeration problems, including maximal chains of non-crossing partitions, labeled trees with marked edges, and factorizations of cyclic permutations.
2. Representation-theoretic properties: Exceptional sequences are central topics in algebraic geometry, representation theory, and combinatorics. For a hereditary algebra $\Lambda$, an exceptional sequence is a sequence $(E_1,\ldots,E_k)$ of rigid indecomposable modules satisfying $\text{Hom}_\Lambda(E_j,E_i) = \text{Ext}_\Lambda(E_j,E_i) = 0$ for all $1 \leq i < j \leq k$.
3. Relative projectivity and injectivity: The authors focus on which elements in exceptional sequences can be relatively projective or relatively injective, an important homological property.

Research Motivation

This paper aims to provide an alternative combinatorial interpretation with representation-theoretic significance:

• Making algebraic concepts visible through planar graphs (rooted labeled forests)
• Establishing a simple one-to-one correspondence between exceptional sequences and rooted labeled forests
• Visualizing the action of the braid group on exceptional sequences

Core Contributions

1. Main bijection theorem: Establishes a bijection between complete exceptional sequences of the linear $A_n$ quiver and rooted labeled forests with $n$ vertices.
2. Characterization of relative projectivity and injectivity: Proves that relatively projective and relatively injective objects in exceptional sequences correspond to descending and ascending vertices in the forest.
3. Complete classification result: Proves that each object in an exceptional sequence of the linearly oriented $A_n$ is either relatively projective or relatively injective (or both).
4. Visualization of braid group action: Constructs a natural action of the braid group on rooted labeled forests and proves its consistency with the known action on exceptional sequences.
5. Generating functions: Provides a three-variable generating function for exceptional sequences, counting the number of relatively projective and relatively injective objects.
6. Action of Garside element: Describes in detail the action of the important Garside element $\Delta$ on rooted labeled forests and its relationship with cluster theory.

Detailed Methodology

Task Definition

Establish a bijection between complete exceptional sequences of the linear $A_n$ quiver $Q: 1 \to 2 \to \cdots \to n$ and rooted labeled forests with $n$ vertices.

Core Construction

1. From Exceptional Sequences to Forests (Hasse Diagram)

Given a complete exceptional sequence $E^* = (E_1,\ldots,E_n)$, define a partial order on the set $\{v_1,v_2,\ldots,v_n\}$: $v_i < v_j \text{ if and only if the support of } E_i \text{ is contained in the support of } E_j$

Theorem 1.13: The Hasse diagram of this partial order is a rooted labeled forest, and conversely, each rooted labeled forest uniquely determines an exceptional sequence.

2. From Forests to Exceptional Sequences (Recursive Construction)

For a rooted labeled forest $F$:

• Let $v_r$ be the root and $v_{i_1},\ldots,v_{i_k}$ be its children
• Order the labels as $i_1 < \cdots < i_p < r < i_{p+1} < \cdots < i_k$
• Assign the module $E_r := M_{1n}$ and construct a filtration to obtain subsequences

Key properties:

• The length of each $E_i$ equals the weight of $v_i$ (subtree size)
• Support containment relations correspond to the partial order in the forest

Technical Innovations

1. Non-crossing Property

Lemma 1.3: The support intervals of any two objects in an exceptional sequence are either disjoint or one contains the other (non-crossing property).

2. Characterization of Relative Projectivity and Injectivity

Theorem 1.21:

• $E_i$ is both relatively projective and relatively injective ⟺ $v_i$ is a root of the forest
• For non-root $v_i$, let $v_j$ be its parent node, then:
  • $E_i$ is relatively projective ⟺ $i < j$ (descending vertex)
  • $E_i$ is relatively injective ⟺ $i > j$ (ascending vertex)

3. Braid Group Action

Define the action of $\sigma_i$ on forests:

• Case 0: When $v_i, v_{i+1}$ are not adjacent, only swap labels
• Case 1: Transformation when $v_i$ is the parent of $v_{i+1}$
• Case 2: Transformation when $v_i$ is a child of $v_{i+1}$
• Case 3: Transformation when $v_i, v_{i+1}$ are siblings

Experimental Setup

Theoretical Verification

The paper is primarily theoretical work, verified through:

1. Small example verification:
   • Case $A_2$: 3 forests cyclically arranged by $\sigma_1$
   • Case $A_3$: Braid group action diagram for 16 forests
2. Generating function verification:
   • Case $A_3$: $P_{A_3}(a,b,c) = 2a^2c + 2b^2c + 5abc + 3ac^2 + 3bc^2 + c^3$
3. Comparison with known results:
   • Comparison with chord diagram construction (Goulden-Yong)
   • Comparison with parking functions

Experimental Results

Main Results

1. Completeness of the Bijection

Proves that the map $H: \{\text{complete exceptional sequences}\} \to \{\text{rooted labeled forests}\}$ is a bijection through construction of a right inverse and counting arguments.

2. Generating Function Formula

Theorem 1.30: $P_n(a,b,c) = c\prod_{j=1}^{n-1}(ja + (n-j)b + c)$ where:

• $a$: number of objects that are only relatively projective
• $b$: number of objects that are only relatively injective
• $c$: number of objects that are both relatively projective and relatively injective

3. Action of Garside Element

Proposition 4.11: $\Delta$ transforms the projective vertices of forest $F$ into the root of $\Delta F$, and transforms the root of $F$ into the injective vertices of $\Delta F$.

Important Findings

1. Complete classification: Each exceptional object of linear $A_n$ is either relatively projective or relatively injective (Corollary 1.22)
2. Visualization of braid group action: Intuitively displays complex braid group transformations through forest diagrams
3. Connection with cluster theory: The Garside element connects support-tilting sets and signed exceptional sequences

Related Work

Major Related Research

1. Exceptional sequence theory: Work by Crawley-Boevey, Ringel and others on braid group actions
2. Combinatorial correspondences: Connections with non-crossing partitions, labeled trees, and cyclic permutation factorizations
3. Chord diagram construction: Goulden-Yong's correspondence between trees and exceptional sequences
4. Parking functions: Alternative combinatorial interpretation of exceptional sequences

Uniqueness of This Work

• First direct bijection between forests and exceptional sequences
• Explicit characterization of relative projectivity/injectivity properties
• New visualization method for braid group actions
• Connection between representation theory and cluster theory

Conclusions and Discussion

Main Conclusions

1. Establishes a perfect correspondence between exceptional sequences of linear $A_n$ and rooted labeled forests
2. Completely characterizes relative projectivity and injectivity through forest structure
3. Achieves geometric visualization of braid group actions
4. Reveals deep connections with cluster theory

Limitations

1. Applies only to linearly oriented quivers of type $A_n$
2. Generalization to other Dynkin types or general quivers remains unresolved
3. Relationship with the parking function approach requires further clarification

Future Directions

1. Generalization to other quiver types
2. Exploration of relationships with bounded-height forests
3. Development of more general representation-theoretic combinatorial correspondences

In-Depth Evaluation

Strengths

1. Strong innovation: Provides a novel combinatorial interpretation of exceptional sequences
2. Deep results: Completely characterizes important homological properties
3. Elegant methods: Reveals complex algebraic relationships through simple forest structures
4. Broad applications: Connects representation theory, combinatorics, and cluster theory

Technical Highlights

1. Utilization of non-crossing property: Cleverly transforms the non-crossing nature of intervals into forest structure
2. Sophistication of recursive construction: Achieves bijection through recursive construction using weights and filtrations
3. Geometrization of braid group action: Transforms abstract braid group actions into intuitive graph transformations

Theoretical Significance

1. Contribution to representation theory: Provides new tools and perspectives for exceptional sequence theory
2. Combinatorial significance: Enriches applications of rooted labeled forests
3. Geometric intuition: Makes abstract algebraic concepts geometrically visualizable

Potential Impact

This work may have significant impact on:

• Research on exceptional sequences in representation theory
• Cluster algebras and tilting theory
• Enumeration problems in combinatorics
• Stability conditions in algebraic geometry

References

This paper cites important literature in the field, including:

• Crawley-Boevey's pioneering work on braid group actions
• Ringel's generalization of exceptional sequences for hereditary algebras
• Goulden-Yong's correspondence between chord diagrams and trees
• Related literature on cluster theory and tilting theory

Overall Assessment: This is a high-quality theoretical paper making important contributions at the intersection of representation theory and combinatorics. By establishing a bijection between exceptional sequences and rooted labeled forests, it not only provides a new combinatorial interpretation but also reveals deep algebraic and geometric structures. The results have significant theoretical value and broad application prospects.