Uniform sum-product phenomenon for algebraic groups and Bremner's conjecture

This paper combines additive combinatorics and Diophantine geometry to establish a uniform sum-product phenomenon for one-dimensional algebraic groups over $\mathbb{C}$, thereby resolving Bremner's conjecture on arithmetic progressions in elliptic curve coordinates and improving upon existing results by Bays--Breuillard regarding Elekes--Szabó-type theorems.

The Gist

Imagine you are a detective trying to solve a mystery about how numbers behave. In the world of mathematics, there are two main "teams" that numbers can join: the Additive Team (where numbers get bigger by adding, like 1, 2, 3, 4...) and the Multiplicative Team (where numbers grow by multiplying, like 1, 2, 4, 8, 16...).

Usually, these two teams don't get along. If a group of numbers is very orderly in the Additive Team (like a perfect arithmetic progression), they tend to be chaotic and messy in the Multiplicative Team, and vice versa. This is called the Sum-Product Phenomenon.

This paper, written by Joseph Harrison, Akshat Mudgal, and Harry Schmidt, is like a master detective report that unifies several different cases where this "team rivalry" causes problems. They use a mix of old-school number theory (Diophantine geometry) and modern pattern-finding (additive combinatorics) to prove some very strong rules about how these numbers must behave.

Here is a breakdown of their three biggest discoveries, explained with everyday analogies:

1. The "Bremner's Conjecture" Case: The Impossible Line-Up

The Mystery: Imagine an elliptic curve (a fancy, curved shape defined by an equation) as a dance floor. The "rational points" are the dancers on this floor. Mathematicians wondered: Can you find a long line of dancers standing in a perfect straight row (an arithmetic progression) on this dance floor?

The Old Belief: People suspected that no matter how big the dance floor is, you can't have an infinitely long straight line of dancers. The length of the line should be limited by how "complex" the dance floor is (specifically, its "rank").

The New Solution: The authors proved this suspicion is correct. They showed that the length of any such line is strictly limited.

• The Analogy: Think of the dance floor as a bumpy, curved trampoline. If you try to lay a perfectly straight, rigid ruler (an arithmetic progression) on it, the ruler will eventually hit a bump and break. The authors calculated exactly how long the ruler can be before it breaks, and they proved this limit depends only on the "bumpiness" of the trampoline, not on the specific size of the ruler.
• Why it matters: This solves a long-standing guess made by a mathematician named Bremner and applies to other patterns too, like geometric progressions (doubling numbers) or sequences of squares.

2. The "Sum-Product" Explosion: The Balloon Effect

The Mystery: If you take a small group of numbers and start adding them together, or multiplying them together, how big does the group get?

• If you have a group that is very "additive" (like a ladder: 1, 2, 3, 4), adding them makes a slightly bigger group, but multiplying them explodes into a huge, chaotic mess.
• If you have a group that is "multiplicative" (like a chain: 2, 4, 8, 16), multiplying them is easy, but adding them creates a huge mess.

The New Solution: The authors proved a "Uniform Bourgain–Chang" estimate. This is a fancy way of saying: You can't be good at both.

• The Analogy: Imagine a balloon. If you try to squeeze it in one direction (make it very orderly in addition), it must bulge out in the other direction (become huge in multiplication). You cannot keep the balloon small and round in both dimensions at the same time.
• The Breakthrough: Previous mathematicians could only prove this for simple number systems (like integers). These authors proved it works for a much wider universe of "algebraic groups" (including complex numbers and elliptic curves). They showed that even if you try to hide the structure, the "bulge" (expansion) is inevitable and massive.

3. The "Elekes–Szabó" Puzzle: The Broken Grid

The Mystery: Imagine you have a giant grid made of points. You draw a curve (a variety) through this grid. If the points on your grid are "clumped together" (they have a small sumset, meaning they are very structured), how many points of the grid can your curve touch?

• If the curve is "degenerate" (it's just a straight line or a simple shape that fits the grid perfectly), it can touch a lot of points.
• If the curve is "non-degenerate" (it's a weird, twisted shape that doesn't fit the grid's rhythm), it should only touch a few points.

The New Solution: The authors proved that if your grid points are clumped together, a "twisted" curve can only touch a tiny fraction of them.

• The Analogy: Imagine a dance floor where everyone is dancing in a perfect circle (the clumped points). Now, imagine a laser beam (the curve) sweeping across the floor. If the laser is straight, it might hit many dancers. But if the laser is bent into a weird, jagged shape that doesn't match the circle, it will only hit a few dancers before missing the rest.
• The Improvement: They didn't just say "it hits fewer points"; they calculated the exact mathematical limit of how few points it can hit, showing that their limit is the best possible one.

How Did They Do It? (The Toolkit)

To solve these mysteries, the authors used a "Swiss Army Knife" of mathematical tools:

1. Old Maps (Diophantine Geometry): They used deep, classical theorems (like Mordell–Lang) that describe how points sit on curves. Think of this as using an old, detailed map of the dance floor to know where the bumps are.
2. New Radar (Additive Combinatorics): They used a very recent, powerful breakthrough (the weak Polynomial Freiman–Ruzsa conjecture) that acts like a high-tech radar, detecting hidden patterns in groups of numbers.
3. The Bridge: The genius of the paper is connecting these two. They showed that the "bumps" on the dance floor (geometry) force the "patterns" in the numbers (combinatorics) to behave in a specific way.

The Bottom Line

This paper is a major step forward in understanding the fundamental rules of numbers. It proves that order and chaos are locked in a tug-of-war. If you try to force numbers to be too orderly in one way (addition), they will inevitably explode in another way (multiplication or geometry).

The authors didn't just solve one puzzle; they built a universal framework that solves many puzzles at once, from the shape of elliptic curves to the behavior of complex polynomials. It's a bit like discovering a new law of physics that explains why a spinning top wobbles, why a balloon pops, and why a dancer can't stand perfectly still on a trampoline—all at the same time.

Technical Summary

Here is a detailed technical summary of the paper "Uniform Sum-Product Phenomenon for Algebraic Groups and Bremner's Conjecture" by Joseph Harrison, Akshat Mudgal, and Harry Schmidt.

1. Problem Statement and Motivation

The paper addresses the fundamental tension between additive and multiplicative (or more generally, group-theoretic) structures within algebraic settings. It unifies three distinct areas of research:

1. Bremner's Conjecture: Concerning the boundedness of the length of arithmetic progressions (and geometric progressions/squares) within the coordinates of rational points on elliptic curves.
2. The Sum-Product Phenomenon: The Erdős–Szemerédi conjecture, which posits that a finite set of numbers cannot be simultaneously additively and multiplicatively structured.
3. Elekes–Szabó Problems: Questions regarding the expansion of sets under polynomial maps and the intersection of varieties with discrete grids.

The authors aim to resolve these problems in a uniform manner over 1-dimensional connected algebraic groups over $\mathbb{C}$ (specifically the additive group $\mathbb{G}_a$, the multiplicative group $\mathbb{G}_m$, and elliptic curves), moving beyond previous results that were often restricted to integers ($\mathbb{Q}$) or relied on non-uniform constants.

2. Methodology

The paper's primary innovation is the synthesis of Diophantine geometry and Additive Combinatorics. The methodology proceeds through three main steps:

A. Reduction to Bounded Rank Subgroups

Using the recent breakthrough resolution of the Weak Polynomial Freiman–Ruzsa (PFR) conjecture over $\mathbb{Z}$ by Gowers, Green, Manners, and Tao, the authors show that any finite set $A$ with small doubling ($|A+A| \le K|A|$) in an algebraic group is essentially contained in a low-rank subgroup (or a bounded number of translates thereof). This allows them to reduce the problem from arbitrary sets to sets lying within subgroups of finite rank $r$.

B. Uniform Mordell–Lang and S-Unit Bounds

Once restricted to finite rank subgroups, the authors apply deep results from Diophantine geometry:

• Uniform Mordell–Lang Conjecture: Utilizing the work of David–Philippon and Laurent, they bound the number of points of a variety intersecting a finite rank subgroup.
• S-Unit Equations: They employ bounds by Evertse–Schlickewei–Schmidt to control the intersection of varieties with subgroups.
  These tools provide effective, uniform bounds on the size of intersections, independent of the specific generators of the subgroup.

C. Non-Degeneracy of Varieties

A critical technical hurdle is ensuring that the varieties constructed to model the sum-product phenomenon are "non-degenerate." The authors define a variety $V \subseteq G^g \times B$ as degenerate if it contains a positive-dimensional translate of an algebraic subgroup. They prove that for correspondences that are not translates of algebraic subgroups, the associated auxiliary varieties (constructed via tangent space analysis and holomorphic maps) are non-degenerate. This ensures that the "structure" of the set is destroyed when mapped through the correspondence, forcing expansion.

3. Key Contributions and Results

A. Resolution of Bremner's Conjecture

The authors prove a uniform bound on the length of arithmetic progressions, geometric progressions, and sequences of squares in the coordinates of rational points on elliptic curves.

• Theorem 1.1: For an elliptic curve $E$ of rank $r$, the size of any such progression $A$ in the coordinates $X$ or $Y$ is bounded by $|A| \le C^{1+r}$, where $C$ is an effectively computable constant independent of the curve's coefficients.
• This generalizes previous partial results and provides a uniform bound for families of curves of bounded rank.

B. Uniform Sum-Product Estimates (Bourgain–Chang Type)

The paper establishes a general unbounded expansion result for 1-dimensional algebraic groups.

• Theorem 1.3: Given integers $k, d$, there exists $g$ such that for any finite set $A$ in a group $G$ (not isomorphic to $\mathbb{G}_a$) and correspondences $C_i$ of degree $d$, either the sumset $|gA|$ or the image sumset $|\sum C_i(A)|$ grows as $|A|^k$.
• Corollary 1.4: This recovers and generalizes the Bourgain–Chang sum-product estimates to sets in $\mathbb{C}$ (not just $\mathbb{Q}$) and allows for complex polynomial coefficients.

C. Improved Elekes–Szabó Bounds

The authors refine the understanding of how varieties intersect with "structured" sets (sets with small doubling).

• Theorem 1.5 & 2.6: They provide an upper bound for $|V \cap A^g|$ that depends on the coset defect (the maximal dimension of a coset of a subgroup contained in $V$).
• The bound is shown to be quantitatively optimal. If $V$ is not a translate of a subgroup, the intersection size is significantly smaller than the trivial bound, specifically $|V \cap A^g| \ll |A|^{\dim(V) - \eta}$.

D. Asymmetric Sum-Product Results

They prove an asymmetric version of the Bays–Breuillard conjecture (Theorem 2.3), showing that for a correspondence $C$, either $|A+A|$ or $|C(A)+C(A)|$ must be large, with the exponent $\delta$ being independent of the specific groups involved (though dependent on the degree of the correspondence).

4. Significance and Impact

• Unification: The paper successfully unifies the study of arithmetic progressions in elliptic curves with the general sum-product phenomenon, demonstrating that they are manifestations of the same underlying incompatibility between algebraic group structures and additive structures.
• Uniformity: Unlike previous works where constants depended on the specific curve or field, these results provide uniform bounds depending only on the rank of the group and the degree of the varieties.
• Methodological Shift: By leveraging the Weak PFR conjecture (a purely additive combinatorial result) to unlock problems in Diophantine geometry (Mordell–Lang), the paper opens a new pathway for solving problems involving rational points on varieties.
• Optimality: The authors demonstrate that their power-saving bounds in the Elekes–Szabó setting are the best possible, settling questions about the precise order of magnitude for these intersections.
• Applications: The results have immediate implications for Diophantine equations, specifically determining the Zariski closure of rational points on certain high-dimensional surfaces (Theorem A.1), confirming that they consist of finite unions of elliptic curves and points.

In summary, this work represents a major advancement in the intersection of number theory and combinatorics, providing powerful, uniform tools to analyze the distribution of rational points and the expansion properties of sets within algebraic groups.