Stable norm (old&new)

TSIMF 10/11/26

  • Greg Mc Shane
    • Institut Fourier
    • USTC, Hefei
Sanya 2026

  • slides : google greg mcshane github
  • click on talk slides
Sanya 2026

Markoff numbers are integers that appear a Markoff triple

which are solutions of a Diophantine equation
the so-called Markoff cubic

Sanya 2026

Odd index Fibonacci numbers are Markoff numbers

Sanya 2026

Odd index Pell numbers are Markoff numbers

Sanya 2026

Frobenius uniqueness conjecture

The largest integer in a triple determines the two other numbers.

Sanya 2026

Partial results

m = biggest Markoff number determines the two other numbers.

Sanya 2026

Martin Aigner

Sanya 2026

Aigner's monotonicity conjectures

  • Markoff’s theorem and 100 years of the uniqueness conjecture. A mathematical journey from irrational numbers to perfect matchings. 2013.
    • M. Rabideau, R. Schiffler,
      Continued fractions and orderings on the Markoff numbers,
      Advances in Mathematics Vol 370, 2020. published
    • C Lagisquet and E. Pelantová and S. Tavenas and L. Vuillon, On the Markoff numbers: fixed numerator, denominator, and sum conjectures. published
Sanya 2026

There is a natural map (we'll see why shortly)

Sanya 2026

Aigner's conjectures proof

Sanya 2026

Labeling Markoff numbers

A tale of three trees

  • Farey "tree" of coprime integers
  • Markoff tree of solutions to the cubic
  • Bass-Serre of a free product
    • automorphisms of the Markoff cubic
    • automorphisms of the fundamental group of the punctured torus

Sanya 2026

coprime integers

  • closed geodesic
  • arc on a punctured torus (disjoint from the geodesic)
Sanya 2026


source

Sanya 2026

source

Sanya 2026

natural map ?

Markoff numbers

  • action on

  • action on Markoff numbers/triples ?

  • Vieta jumping

    • quadratic in , two roots
    • Vieta formula
Sanya 2026

Automorphisms

Vieta jumping/ flips
(cyclic) permutations of

action of

Sanya 2026

Natural = -equivariant map

  • actions = projective on left and by autos on right
Sanya 2026

  • Uniqueness conjecture
  • The largest integer in a triple determines the two other numbers.
  • The multiplicity of any number in the complementary regions to the tree is at most 6

M., Parlier 2007:

Sanya 2026

Geometry of Aigner's conjectures

  • is the simple closed geodesic on the punctured torus
  • is in the homology class
  • is the stable norm of the homology class
Sanya 2026

Geometric theory: H. Cohn

Approach to Markoff’s Minimal Forms Through Modular Functions (1955)

  • modular torus = quotient of upper half plane by commutator subgroup of , acting by Mobius transformations
  • (essentially) relates Markoff numbers to lengths of simple closed geodesics
Sanya 2026

  • modular torus = quotient of upper half plane by commutator subgroup of
  • obtained from a pair of ideal triangles by identification
Sanya 2026

Character variety

modular torus =

  • fundamental group of the torus.
  • any hyperbolic torus = ,
  • discrete faithful representation
  • lifts to
  • generators of
    • Definition: character map
Sanya 2026

Theorem: Fricke, Cohn (and others)

The semi-algebraic set:

can be identified with the Teichmueller space of the punctured torus.

  • permutations
  • the Vieta flips

used to construct Markoff's binary tree are induced by automorphisms of the fundamental group of the torus.

Sanya 2026

Counting problem

Sanya 2026

Counting closed simple geodesics

  • character map
  • is generator iff essential simple closed curve representing its conjugacy/free homotopy class
Sanya 2026

Simple representatives

  • blue curve is simple representative of its homotopy class
  • not every homotopy class contains a simple curve
  • every (non trivial) homology class has a representative that is a (multiple) of a simple curve
Sanya 2026

Norms and minimizers

Let be an essential closed curve its length.

  • any pair of curves in linearly independent homology classes intersect
    • a curve with self intersections is never a minimizer
    • convexity/triangle inequality
Sanya 2026

  • Cluster algebra folks call this a smoothing
Sanya 2026

Corollary: Let T be a punctured torus with a hyperbolic structure.

  • Then, the shortest multicurve representing a non-trivial homology class is a simple closed geodesic if is a primitive homology class, and a multiply covered geodesic otherwise.
  • In addition, the shortest multicurve representing is unique.
Sanya 2026

Unit ball

  • are a pair of shortest closed geodesics on the
    modular torus.
  • the homology class of
    • is
    • is
Sanya 2026

Unit ball and counting

  • is the probability that 2 random integers are coprime.
  • the area of the unit ball depends on the hyperbolic structure
  • with Rivin we studied it, but now it's called the Mirzakhani function :(
Sanya 2026

Why log ?

  • important monotone increasing on
Sanya 2026

Aigner's conjectures

Sanya 2026

Reformulate Aigner's conjectures

Markoff number
monotone increasing on

  • Let be real non negative numbers and then

  • If in addition then

Sanya 2026

Aigner's conjectures proof

Sanya 2026

Conjecture

Sanya 2026

How do you draw the norm ball?

  • Plot

Sanya 2026

Klempnauer and Schröder: “The stable norm on the 2-torus at
irrational directions”.
Nonlinearity: 30.3 (2017)

Sanya 2026

  • On the stable norm of slit tori and the Farey sequence
    Pablo Montealegre. arXiv: 2310.05570
Sanya 2026

How do you draw the norm ball?

  • Plot

  • corners at rational directions
  • smoothness at irrational directions
Sanya 2026

Gauss-Bonnet

For a polygon in the Euclidean plane

  • interior angles
  • geodesic curvature of the edges
Sanya 2026

Gauss-Bonnet

  • For the norm ball:
    • zero curvature at irrational directions
    • corners at rational directions, identity

Sanya 2026

Norm balls for different (singular) metrics on the punctured torus.

monodromy
<0 geodesic boundary hyperbolic
=0 puncture parabolic
>0 cone point elliptic
Sanya 2026

Calculating the angles

    • slopes of support lines to the norm ball
  • pick so that
  • consider in the homology class

Sanya 2026

Igor Rivin

  • Banned from the ENS Lyon
  • Banned from the IHES
  • Tried to get a job by buying Gromov a dog
Sanya 2026

Igor Rivin

  • me: sure it's complicated but I can probably integrate out some of
    the variables
  • Igor: yeah, but you'll probably get the same identity back

Sanya 2026

How do you draw the norm ball?

  • Plot

  • corners at rational directions
  • smoothness at irrational directions
Sanya 2026

Robert Hines' perimeter formula

  • An infinite product on the Teichmüller space of the once-punctured torus where is a positive integer "height"

  • is the length of the geodesic,
  • are the lengths of any triple of simple geodesics intersecting at a single point.
  • it's a norm!!!!
Sanya 2026

Analysis of (the derivative of) F

  • is a continuous extension of

  • projection of the graph of the "inverse" of the norm ball
    coordinates
  • Fock and Goncharov. Dual Teichmuller and lamination spaces. Handbook of Teichmuller theory. Vol. I
Sanya 2026

Igor Rivin

  • me: sure it's complicated but I can probably integrate out some of
    the variables
  • Igor : yeah, but you'll probably get the same identity back

Sanya 2026

How do you draw the norm ball?

  • Plot

  • corners at rational directions
  • smoothness at irrational directions
Sanya 2026

Robert Hines' perimeter formula

  • An infinite product on the Teichmüller space of the once-punctured torus where is a positive integer "height"

  • is the length of the geodesic,
  • are the lengths of any triple of simple geodesics intersecting at a single point.

  • it's a norm!!!!
Sanya 2026

Analysis of (the derivative of) F

  • is a continuous extension of

  • projection of the graph of the "inverse" of the norm ball
    coordinates
  • Fock and Goncharov. Dual Teichmuller and lamination spaces. Handbook of Teichmuller theory. Vol. I
Sanya 2026

Igor Rivin

In an appendix Hines gives a variation of his argument and reproves

"confirming" Rivin's intuition

  • me: sure it's complicated but I can probably ....
  • Igor : yeah, but you'll probably get the same identity back
Sanya 2026

#

## infinity of Markoff triples: $z=1$

$\begin{pmatrix} 3 & -1 \\ 1 & 0 \end{pmatrix}$

is an automorph of

$$x^2 + y^2 - 3x y.$$

So $( v_n,v_{n+1},1)$ is a Markoff triple where

$\begin{pmatrix} x \\ y \end{pmatrix}= \begin{pmatrix}v_{n+1} \\ v_n \end{pmatrix} = \begin{pmatrix} 3 & -1 \\ 1 & 0 \end{pmatrix}^n \begin{pmatrix}1 \\ 1 \end{pmatrix}$

* snake graph and its perfect matchings

* "lengths" that verify a Ptolemy inequality

#

## Group actions

<!-1- _transition: glow -1->

$\mathbb{Q}\cup \infty \subset$ circle/projective line

* $(a,c)\text{ primitive } \mapsto a/c \in \mathbb{Q}\cup \infty$

* $\begin{pmatrix} a & d \\ c & d \end{pmatrix} \mapsto$ arc joining $(a/c, b/d)$

* $(a/b, c/d)$ are Farey neighbors iff $|ad - bc | = 1$

* obvious transitive $SL(2,\mathbb{Z})$ action on Farey neighbors

#

![width:600px](./pozzi.jpg.png)

[source](https://www.mathi.uni-heidelberg.de/~pozzetti/trees/4.pdf)

* involution $(x^-,y,z) \mapsto (x^+, y,z) = (3yz - x^-, z,y)$

* elliptic involution swaps triangles fixes midpoint of diagonal

#### Exo

- Nielsen move $\rightarrow$ Vieta flip

- $tr\,ab + tr\,ab^{-1} = (tr\,a) (tr\,b)$

#

### Simple representatives in homology

$\phi : \mathbb{Z}*\mathbb{Z} \rightarrow \mathbb{Z}^2 \simeq

H^1(T,\mathbb{Z})$.

abelianizing homomorphism.

- $\phi$ takes generators of $\mathbb{Z}*\mathbb{Z}$ to generators of $\mathbb{Z}^2$.

- $(p,q) \in \mathbb{Z}^2$ generator $\Leftrightarrow p,q$ coprime.

#

![w:800](./minimizer.png)

- $m_{p/q} = \frac13 tr \hat{\rho}( \gamma_{p/q})$