Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

The (noncommutative) geometry of difference equations

Eric M. Rains

TL;DR

The work develops a comprehensive geometric framework in which difference and differential equations with singularity data are realized as (derived) sheaves on noncommutative, and in particular quasi-ruled, surfaces. By translating equations into sheaf data via anticanonical curves, spectral curves, and Higgs-type structures, the authors build a robust moduli theory for these objects, including projectivity and Poisson/symplectic foliations on moduli spaces. A central technical achievement is the construction and control of minimal lifts and invisible sheaves under birational transformations, together with a derived-category perspective that yields derived equivalences and Lagrangian structures on moduli, thereby connecting discrete isomonodromy, Painlevé-type dynamics, and Sakai-type surfaces within a unified Poisson-geometric, birational framework. The results culminate in a detailed interplay between commutative relaxations on rational/ruled surfaces and noncommutative deformations, enabling a principled understanding of how Lax pairs, isomonodromy, and moduli of equations behave under birational changes and line-bundle twists. This provides a foundation for future derived-equivalence statements linking differential and difference equations and points toward a noncommutative, modular theory of integrable systems with broad mathematical and potentially physical significance.

Abstract

The aim of this monograph is twofold: to explain various nonautonomous integrable systems (discrete Painlevé all the way up to the elliptic level, as well as generalizations à la Garnier) using an interpretation of difference and differential equations as sheaves on noncommutative projective surfaces, and to develop the theory of such surfaces enough to allow one to apply the usual GIT construction of moduli spaces of sheaves. This requires a fairly extensive development of the theory of birationally ruled noncommutative projective surfaces, both showing that the analogues of Cremona transformations work and understanding effective, nef, and ample divisor classes. This combines arXiv:1307.4032, arXiv:1307.4033, arXiv:1907.11301, as well as those portions of arXiv:1607.08876 needed to make things self-contained. Some additional results appear, most notably a proof that the resulting discrete actions on moduli spaces of equations are algebraically integrable.

The (noncommutative) geometry of difference equations

TL;DR

The work develops a comprehensive geometric framework in which difference and differential equations with singularity data are realized as (derived) sheaves on noncommutative, and in particular quasi-ruled, surfaces. By translating equations into sheaf data via anticanonical curves, spectral curves, and Higgs-type structures, the authors build a robust moduli theory for these objects, including projectivity and Poisson/symplectic foliations on moduli spaces. A central technical achievement is the construction and control of minimal lifts and invisible sheaves under birational transformations, together with a derived-category perspective that yields derived equivalences and Lagrangian structures on moduli, thereby connecting discrete isomonodromy, Painlevé-type dynamics, and Sakai-type surfaces within a unified Poisson-geometric, birational framework. The results culminate in a detailed interplay between commutative relaxations on rational/ruled surfaces and noncommutative deformations, enabling a principled understanding of how Lax pairs, isomonodromy, and moduli of equations behave under birational changes and line-bundle twists. This provides a foundation for future derived-equivalence statements linking differential and difference equations and points toward a noncommutative, modular theory of integrable systems with broad mathematical and potentially physical significance.

Abstract

The aim of this monograph is twofold: to explain various nonautonomous integrable systems (discrete Painlevé all the way up to the elliptic level, as well as generalizations à la Garnier) using an interpretation of difference and differential equations as sheaves on noncommutative projective surfaces, and to develop the theory of such surfaces enough to allow one to apply the usual GIT construction of moduli spaces of sheaves. This requires a fairly extensive development of the theory of birationally ruled noncommutative projective surfaces, both showing that the analogues of Cremona transformations work and understanding effective, nef, and ample divisor classes. This combines arXiv:1307.4032, arXiv:1307.4033, arXiv:1907.11301, as well as those portions of arXiv:1607.08876 needed to make things self-contained. Some additional results appear, most notably a proof that the resulting discrete actions on moduli spaces of equations are algebraically integrable.

Paper Structure

This paper contains 113 sections, 383 theorems, 1127 equations, 6 figures.

Table of Contents

  1. Introduction
  2. Acknowledgements
  3. Commutative geometry
  4. Sheaves from difference equations
  5. Symmetric elliptic equations
  6. Degenerations and generalizations
  7. The spectral curve
  8. Isospectral transformations
  9. Quadratic and other transformations
  10. Birational classification of Poisson surfaces
  11. Poisson structures
  12. Poisson surfaces
  13. Lifting through birational morphisms
  14. The minimal lift
  15. Invisible sheaves
  16. ...and 98 more sections

Key Result

Proposition 3.1

Let $L/K$ be a quadratic field extension, and let $A\in \mathop{\mathrm{GL}}\nolimits_n(L)$ be a matrix such that $\bar{A}=A^{-1}$, where $\bar{\cdot}$ is the conjugation of $L$ over $K$. Then there exists a matrix $B\in \mathop{\mathrm{GL}}\nolimits_n(L)$ such that $A = \bar{B} B^{-1}$, and $B$ is

Figures (6)

  • Figure 1: The natural stratification of ${\cal X}_7^{\alpha,\ge -2}/W(E_8)$ over $\mathbb Z[1/6]$.
  • Figure 2: The naïve poset of types in ${\cal X}_7^{\alpha,\ge -2}/W(E_8)$.
  • Figure 3: The poset of hypergeometric functions generalizing ${}_2F_1$.
  • Figure 4: The number of types of equation for each type of hypergeometric function
  • Figure 5: The poset of hypergeometric sums of type ${}_2F_1$
  • ...and 1 more figures

Theorems & Definitions (961)

  • Proposition 3.1
  • proof
  • Proposition 3.2
  • proof
  • Remark
  • Remark
  • Remark
  • Proposition 3.3
  • proof
  • Remark 1
  • ...and 951 more