Bumpless pipe dreams meet Puzzles

Basic Information

• Paper ID: 2309.00467
• Title: Bumpless pipe dreams meet Puzzles
• Authors: Neil J.Y. Fan, Peter L. Guo, Rui Xiong
• Classification: math.CO math.AG
• Publication Date: September 1, 2023
• Paper Link: https://arxiv.org/abs/2309.00467

Abstract

This paper establishes a triple Schubert calculus version of the product expansion of double Grothendieck polynomials, namely an expansion formula for the product of double Grothendieck polynomials $G_u(x,y) \cdot G_v(x,t)$ indexed by permutations with separated descents. The formula is expressed in the form of "pipe puzzles," which integrates the structures of bumpless pipe dreams and classical puzzles. As direct applications, this formula recovers the separated descent puzzle formula of Knutson-Zinn-Justin and the bumpless pipe dream model of Weigandt for double Grothendieck polynomials, and partially confirms Kirillov's positivity conjecture concerning skew operators on Schubert polynomials.

Research Background and Motivation

Core Problem

The core problem addressed in this paper is the computation of structure constants in triple Schubert calculus, specifically the expansion of the product: $G_u(x,y) \cdot G_v(x,t) = \sum_w c^w_{u,v}(t,y) \cdot G_w(x,t)$ where $u,v$ are permutations with separated descents: $\max \text{des}(u) \leq k \leq \min \text{des}(v)$.

Significance

1. Geometric Significance: This corresponds to triple Schubert calculus in the equivariant K-theory of the ring of flag varieties, possessing important algebraic-geometric implications
2. Combinatorial Significance: Unifies multiple important combinatorial models, including classical puzzles and bumpless pipe dreams
3. Theoretical Value: Provides explicit combinatorial formulas for polynomial expansions, facilitating understanding of the structure of Schubert polynomials

Limitations of Existing Methods

• The Knutson-Zinn-Justin formula applies only to the case $y=t$
• The Wheeler-Zinn-Justin formula is restricted to Grassmannian permutations
• Lack of a unified combinatorial framework for handling cases with different secondary variables

Core Contributions

1. Main Theorem: Establishes the pipe puzzle formula for separated descent permutations (Theorem 2.5), providing a combinatorial expression for the coefficients $c^w_{u,v}(t,y)$
2. Technical Innovation: Introduces the concept of pipe puzzles, integrating the structures of bumpless pipe dreams and classical puzzles
3. Methodological Breakthrough: Discovers two classes of recurrence relations, simplifying the proof process
4. Application Extension: Recovers multiple known results and confirms Kirillov's positivity conjecture
5. Integrable Model: Realizes pipe puzzles as integrable lattice models satisfying the Yang-Baxter equation

Detailed Methodology

Task Definition

Given permutations $u,v \in S_n$ with separated descents: $\max \text{des}(u) \leq k \leq \min \text{des}(v)$ Compute the coefficients of the product expansion of double Grothendieck polynomials: $G_u(x,y) \cdot G_v(x,t) = \sum_w c^w_{u,v}(t,y) \cdot G_w(x,t)$

Model Architecture

1. Boundary Markings

Define boundary markings on an $n \times n$ grid:

• Right boundary: $\kappa^u_i = \begin{cases} u^{-1}(i), & u^{-1}(i) \leq k \\ 0, & u^{-1}(i) > k \end{cases}$
• Top boundary: $\theta^v_i = \begin{cases} 0, & v^{-1}(i) \leq k \\ v^{-1}(i), & v^{-1}(i) > k \end{cases}$
• Bottom boundary: $\eta^w_i = w^{-1}(i)$

2. Available Tiles

Pipe puzzles employ the following tiles:

• Empty tile: $\square$
• Elbow tiles: $\lceil$, $\lfloor$, $\ulcorner$, $\urcorner$
• Straight pipe tiles: $|$, $-$
• Collision tile: $\times$ (new)

3. Constraints

• Elbow Constraint: Horizontal pipes must receive smaller labels
• Collision Constraint:
  • Same-side pipes: Northwest pipe receives larger label
  • Different-side pipes: Northwest pipe must originate from the right side

Weight Computation

Weight of each tile at position $(i,j)$:

1. Empty tile: $t_j \ominus y_i$
2. Elbow tile (right-side pipe): $1 + \beta(t_j \ominus y_i)$
3. Elbow tile (top-side pipe): $1 + \beta(t_j \ominus y_i)$
4. Collision tile (same-side): $\beta$
5. Collision tile (different-side): $\beta(1 + \beta(t_j \ominus y_i))$
6. Other tiles: $1$

where $x \ominus y = \frac{x-y}{1+\beta y}$.

Technical Innovation Points

1. Recurrence Relations

Discovers two classes of key recurrence relations:

• Recurrence with respect to $u$ (Proposition 3.1): When $s_i u < u$$c^w_{s_i u,v} = -\frac{1+\beta y_i}{y_i - y_{i+1}} c^w_{u,v} + \frac{1+\beta y_{i+1}}{y_i - y_{i+1}} c^w_{u,v}|_{y_i \leftrightarrow y_{i+1}}$
• Recurrence with respect to $w$ (Proposition 3.2): When $s_i w > w$, different forms depending on the relationship between $s_i v$ and $v$

2. Integrability

Through two classes of R-matrices $R^{\text{row}}$ and $R^{\text{col}}$, proves that the lattice model satisfies the Yang-Baxter equation, ensuring the integrability of the model.

Experimental Setup

Mathematical Verification

This paper primarily employs mathematical proof for verification, including:

1. Verification of recurrence relations
2. Confirmation of initial conditions
3. Verification of the Yang-Baxter equation

Specific Examples

• Example 2.4: The case $u = 42135, v = 14532, k = 2$
• Example 2.6: Demonstrates nine pipe puzzles and their weight computations

Experimental Results

Main Results

Theorem 2.5 (Main Result): For permutations $u,v \in S_n$ with separated descents, $c^w_{u,v}(t,y) = \sum_{\pi \in PP(u,v,w)} \text{wt}(\pi)$

Verification of Special Cases

1. Case $y = t$: Recovers the Knutson-Zinn-Justin formula
2. Case $\beta = 0$: Obtains the Schubert polynomial version (Theorem 2.2)
3. Case $v = \text{id}, x = t$: Recovers Weigandt's bumpless pipe dream model

Application Results

• Kirillov Conjecture: For permutations with separated descents, confirms that $\partial_{w/v}S_u(x) \in \mathbb{Z}_{\geq 0}[x_1,x_2,\ldots]$

Related Work

Major Related Research

1. Knutson-Tao: Foundational work on classical puzzle formulas
2. Knutson-Zinn-Justin: Puzzle formulas for separated descent cases
3. Wheeler-Zinn-Justin: Puzzle formulas for Grothendieck polynomials
4. Weigandt: Bumpless pipe dream model
5. Lam-Lee-Shimozono: Bumpless pipe dreams for Schubert polynomials

Contributions of This Paper

Compared to existing work, this paper:

• Unifies multiple known models
• Handles more general cases (different secondary variables)
• Provides an integrable lattice model realization

Conclusions and Discussion

Main Conclusions

1. Successfully establishes the pipe puzzle formula for triple Schubert calculus
2. Proves the integrability of this formula
3. Unifies multiple important combinatorial models
4. Partially resolves Kirillov's positivity conjecture

Limitations

1. Applies only to permutations with separated descents
2. Further research needed for the general case of arbitrary permutations
3. Computational complexity may be high

Future Directions

1. Extension to more general permutation classes
2. Investigation of computational efficiency optimization
3. Exploration of connections with other algebraic structures

In-Depth Evaluation

Strengths

1. Theoretical Depth: Provides profound mathematical insights, connecting multiple important combinatorial structures
2. Technical Innovation: The concept of pipe puzzles is original, cleverly integrating different combinatorial models
3. Proof Techniques: Simplifies complex proofs through recurrence relations and integrable model methods
4. Application Value: Recovers multiple known results and resolves open problems

Weaknesses

1. Scope of Applicability: The separated descent condition is quite restrictive
2. Computational Complexity: Computational efficiency for large-scale problems is insufficiently discussed
3. Geometric Interpretation: Lacks in-depth exposition of the geometric significance of pipe puzzles

Impact

1. Academic Contribution: Provides new combinatorial tools for Schubert calculus
2. Methodological Value: The methods of recurrence relations and integrable models have universal applicability
3. Subsequent Research: Lays the foundation for further research in related fields

Applicable Scenarios

• Algebraic combinatorics research
• Computations in Schubert calculus
• Symmetric function theory
• Integrable systems research

References

The paper cites 23 important references, covering classical works in Schubert calculus, combinatorics, integrable systems and other fields, reflecting the breadth and depth of the research.