🔢 mathematics

The Hessian of elliptic curves as a Lattès map

This paper establishes that the Hessian transformation of elliptic curves is a rigid Lattès map lifting to a degree-3 endomorphism, thereby providing a complete characterization of its dynamics and functional graphs over arbitrary fields (including finite fields) through the action of this endomorphism on the twists of the underlying curve.

Original authors: Marzio Mula, Federico Pintore, Daniele Taufer

Published 2026-08-03

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Marzio Mula, Federico Pintore, Daniele Taufer

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a world where numbers aren't just static symbols on a page, but characters in a bustling, invisible city. In this city, every number has a home, and there are secret rules that dictate how they move from one place to another. This is the realm of dynamical systems, a branch of mathematics that studies what happens when you keep applying the same rule over and over again. Think of it like a game of "telephone" played with numbers: you start with a value, apply a transformation, get a new value, apply the rule again, and watch the story unfold. Sometimes the numbers get stuck in a loop, running in circles forever; other times they wander off into the distance, never to return.

The specific "game" this paper investigates involves elliptic curves. Don't let the name scare you; in this context, an elliptic curve is just a special kind of smooth, curvy shape drawn on a grid. These shapes are famous in math because they have a hidden superpower: you can add points on the curve together, just like adding regular numbers, following a very specific set of instructions. The paper focuses on a particular move called the Hessian transformation. You can think of this as a magical lens that looks at a curve and spits out a new, slightly different curve. If you keep using this lens, you get a chain of curves, creating a complex map of where the numbers go. Mathematicians care about this because understanding these maps helps us solve deep puzzles about numbers, and surprisingly, these patterns might one day help us build unbreakable codes to protect our digital secrets.

So, what did the authors of this paper actually discover? They took a closer look at the Hessian transformation and realized it wasn't just a random shuffle of numbers. They proved that this transformation is actually a very rigid, highly structured type of map known as a Lattès map. To understand what that means, imagine the elliptic curve as a giant, perfect factory floor where workers (the points on the curve) move in a very orderly, symmetrical way. The Hessian transformation is like a camera taking a photo of this factory floor, but it squashes the 3D movement down into a 2D picture. The authors showed that this "squashing" is so precise that the chaotic-looking 2D picture is actually just a shadow of a much simpler, more predictable 3D dance happening on the factory floor.

The paper's main finding is that we can completely predict the behavior of this Hessian game by studying the underlying factory dance. They identified a specific "dance move" (a mathematical operation called an endomorphism) that happens on the curve itself. By understanding this move, they could explain exactly why some numbers get stuck in loops, why others wander off, and how the whole system is connected. They didn't just guess; they built a mathematical proof showing that the Hessian map fits into a specific, rigid framework. This means that instead of trying to guess where a number will go next, we can calculate it with certainty by looking at the "twists" and "turns" of the underlying curve.

The authors also tackled the question of what happens when we play this game with a limited set of numbers, like in a computer system that uses a finite field (think of a clock that only has 12 hours, or a specific number of days in a year). They found that the structure of the game changes depending on the size of this "clock." If the number of hours is divisible by 3 in a certain way, the game breaks down into six distinct, known patterns. If it's not, the game merges two identical patterns into one big, orderly structure. Because they understood the rules so well, they were able to create a fast, efficient method to calculate where any number will end up after many, many steps. This isn't just a theoretical curiosity; the paper suggests that because these patterns are so predictable and structured, they could be useful for cryptography—the science of making secret codes. By knowing exactly how these numbers behave, we might be able to design better locks for our digital world, or perhaps even find new ways to break old ones. The paper doesn't claim to have solved every mystery of the universe, but it has definitely mapped out the streets of this particular mathematical neighborhood with incredible clarity.

Technical Summary: The Hessian of Elliptic Curves as a Lattès Map

Problem Statement
The paper investigates the discrete dynamical systems generated by the Hessian transformation of elliptic curves. Given a projective plane cubic $V(F)$ , its Hessian is the zero locus of the determinant of the Hessian matrix of $F$ . When $V(F)$ is a smooth cubic (an elliptic curve), the Hessian defines a new cubic, inducing a transformation on the moduli space of elliptic curves. The authors focus on two specific projective maps:

$H: \mathbb{P}^1(k) \to \mathbb{P}^1(k)$ , acting on $j$ -invariants.
$\Lambda: \mathbb{P}^1(k) \to \mathbb{P}^1(k)$ , acting on the Hesse pencil (curves of the form $x^3+y^3+z^3+\lambda xyz$ ).

The central problem is to characterize the structure and symmetries of the functional graphs (Hessian graphs) of these maps over arbitrary fields $k$ (with $\text{char}(k) \neq 2, 3$ ) and, specifically, over finite fields $\mathbb{F}_q$ . While the dynamics of rational functions are a broad field of study, the specific structure of Hessian graphs had not been fully investigated, particularly regarding their symmetries and the lifting of these dynamics to elliptic curves.

Methodology
The authors employ a geometric and arithmetic approach, reframing the Hessian transformation as a Lattès map. A Lattès map is a rational function $\phi: \mathbb{P}^1(k) \to \mathbb{P}^1(k)$ that fits into a commutative diagram involving an elliptic curve $E$ , an endomorphism $\psi: E \to E$ , and a finite projection $\pi: E \to \mathbb{P}^1$ .

The methodology proceeds through the following steps:

Lattès Identification: The authors prove that both $H$ and $\Lambda$ are rigid Lattès maps. They explicitly construct a reduced Lattès diagram where the underlying curve is $E: y^2 = x^3 - 1728$ (a curve with $j$ -invariant 0). The lifted map $\psi$ is identified as a degree-3 endomorphism corresponding to complex multiplication by $\sqrt{-3}$ .
Group-Theoretic Analysis: The paper analyzes the dynamics of group endomorphisms with prime kernels (specifically kernel size 3). It classifies the connected components of such functional graphs into cycles, oriented lines, or oriented semilines, with vertices forming regular arborescences (rooted trees).
Projection Analysis: A significant portion of the work is dedicated to understanding how the regular dynamics of $\psi$ on the elliptic curve $E$ project onto the functional graph of the Lattès map $\phi$ via $\pi$ . The authors characterize the "fibers" of the projection, showing they are unions of twists of the elliptic curve. They determine exactly which branches of the ancestor trees are identified under the projection, particularly at points with zero coordinates (torsion points).
Arithmetic Specialization: The results are specialized to finite fields $\mathbb{F}_q$ . The authors utilize the trace of the elliptic curve and the parity of the number of rational points to determine the structure of the fibers and the resulting graph topology.

Key Contributions and Results

Rigid Lattès Structure: The paper establishes that the Hessian transformation is a rigid Lattès map (Theorem 3.19). Unlike flexible Lattès maps (which arise from multiplication-by- $m$ maps), these maps arise from endomorphisms of curves with Complex Multiplication (CM), specifically $E$ with $j=0$ .
Inverse Hessian and 2-Torsion: A bijection is established between the inverse image $H^{-1}(j)$ and the set of $k$ -rational points of order 2 on the elliptic curve with $j$ -invariant $j$ (Proposition 2.13). This implies that computing the inverse Hessian is equivalent to finding 2-torsion points.
Structural Classification of Graphs:
- General Fields: The connected components of the Hessian graphs are shown to be derived from the action of $\psi$ on the twists of $E$ . The projection $\pi$ identifies specific branches of the ancestor trees, particularly those arriving at 2-torsion points and, for the map $H$ , points $\pm T_k$ (3-torsion points).
- Finite Fields: The paper provides a complete classification of Hessian graphs over $\mathbb{F}_q$ $F_{q}$ .
  - If $q \equiv 1 \pmod 3$ , the graphs consist of unions of six known functional graphs of group endomorphisms, pruned at specific branches (Theorem 7.8).
  - If $q \equiv 2 \pmod 3$ , the curve $E$ is supersingular. The graphs arise from an ordered merge of two isomorphic graphs, with depth and cycles fully determined (Theorem 7.13).
Leaf Characterization: The paper characterizes the "leaves" of the Hessian graph (elements with empty preimages). Over finite fields, an element $j$ is a leaf if and only if the trace of the corresponding elliptic curve is even (Lemma 7.4).
Algorithmic Efficiency: The authors demonstrate that any iteration of the Hessian transformation over $\mathbb{F}_q$ can be computed in polynomial time relative to the size of $q$ (Proposition 8.1).

Significance and Claims
The paper claims to provide an "effective tool" for investigating the dynamics of the Hessian transformation by lifting the problem to the arithmetic of a specific elliptic curve. The primary significance lies in:

Unifying Framework: It unifies the study of Hessian dynamics under the theory of Lattès maps, leveraging the well-understood structure of endomorphisms on elliptic curves.
Complete Classification: It offers a complete structural description of Hessian functional graphs over finite fields, a result not previously available in the literature.
Computational Utility: By reducing the iteration of the Hessian map to operations on the underlying elliptic curve, the paper provides an efficient method for computation.
Cryptographic Potential: The authors note that these results pave the way for potential cryptographic applications. Specifically, they prove that supersingular $j$ -invariants lie in prescribed connected components with specific trace properties modulo 3, suggesting these structures could be relevant for cryptographic protocols involving isogenies or dynamical systems.

The paper maintains a modest tone regarding applications, stating that the results "pave the way" for cryptographic use rather than asserting immediate deployment, and focuses primarily on the theoretical characterization of the dynamical systems.

More like this