🔢 mathematics

Partition regularity of generalized Pythagorean pairs

This paper establishes partition regularity results for homogeneous quadratic equations of the form $ax^2+by^2=cz^2$ under finite colorings of positive integers, demonstrating that monochromatic solutions exist for specific variable pairs by leveraging new uniformity properties of aperiodic multiplicative functions and extending previous arguments on the Pythagorean equation.

Original authors: Nikos Frantzikinakis, Oleksiy Klurman, Joel Moreira

Published 2026-07-28

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Nikos Frantzikinakis, Oleksiy Klurman, Joel Moreira

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 giant, infinite party where every guest is a positive whole number: 1, 2, 3, and so on, stretching out forever. Now, imagine you are the host, and you decide to hand out colored wristbands to everyone. You have a limited number of colors—say, red, blue, and green—and you give them out in a pattern that might seem random or chaotic. The big question in this corner of mathematics, known as arithmetic Ramsey theory, is: no matter how you hand out those colors, can you always guarantee that you'll find a specific group of guests who all share the same color and also fit together to solve a particular math puzzle?

The puzzle in question here is a classic one, but with a twist. You've probably heard of the Pythagorean theorem, where $x^2 + y^2 = z^2$ describes the sides of a right-angled triangle. Mathematicians have long wondered if, in any colored party, you can always find three numbers of the same color that solve this equation. This paper dives into a more general version of that puzzle: $ax^2 + by^2 = cz^2$ , where $a$ , $b$ , and $c$ are just different integer coefficients. The goal is to see if we can always find two numbers, $x$ and $y$ , that share a color and solve the equation, even if the third number, $z$ , is a different color. It's like asking if you can always find two friends wearing matching shirts who can team up to build a specific structure, regardless of what the third person is wearing.

This paper, written by Nikos Frantzikinakis, Oleksiy Klurman, and Joel Moreira, tackles the conditions under which these "generalized Pythagorean pairs" are guaranteed to exist. The authors don't just say "yes" or "no"; they map out exactly when the answer is "yes" and when it depends on other big, unsolved mysteries in math. They prove that if the coefficients $a$ , $b$ , and $c$ meet certain specific criteria (like if the product of $a$ and $c$ is a perfect square), then you are guaranteed to find your matching pair $x$ and $y$ . However, for other cases, they show that the answer is likely "yes," but it relies on a famous unproven guess called an "Elliott-type conjecture." If that guess turns out to be true, then the rule holds for almost all cases. They also carefully explain why some specific combinations of numbers simply won't work, ruling out the idea that any set of coefficients would guarantee a solution.

The authors use a mix of powerful mathematical tools to get their answers. They treat the numbers like a landscape, looking for "concentrations" where certain patterns cluster together. They also use a strategy similar to sorting a massive deck of cards into "regular" and "irregular" piles. The "regular" cards (called pretentious functions) behave predictably, while the "irregular" ones (aperiodic functions) are chaotic and cancel each other out. By proving that the chaotic parts disappear and the predictable parts line up just right, they show that the matching pairs must exist.

In short, this paper is a major step forward in understanding these number puzzles. It confirms that for many specific setups, the universe of numbers is friendly enough to always provide a matching pair. For the trickier setups, it provides a conditional "yes," saying, "If the math world behaves the way we think it does, then the answer is yes." It also draws a clear line in the sand, showing exactly which setups are impossible, ensuring we don't waste time looking for solutions where none exist. The work builds on previous discoveries about the classic Pythagorean equation but expands the horizon to a much wider, more complex world of quadratic equations.

Technical Summary: Partition Regularity of Generalized Pythagorean Pairs

1. Problem Statement and Context

The paper addresses a central problem in arithmetic Ramsey theory: determining the conditions under which a homogeneous quadratic equation is partition regular. An equation is partition regular if, for any finite coloring of the positive integers $\mathbb{N}$ , there exists a solution where all variables share the same color.

The specific focus is on equations of the form:
$ax^2 + by^2 = cz^2$
where $a, b, c \in \mathbb{Z} \setminus \{0\}$ . While the full partition regularity of the Pythagorean equation ( $x^2 + y^2 = z^2$ ) remains an open problem, the authors investigate pairwise partition regularity.

Definition: The equation is partition regular with respect to a pair of variables (e.g., $x, y$ ) if, for every finite coloring of $\mathbb{N}$ , there exist distinct $x, y$ of the same color and some $z \in \mathbb{N}$ satisfying the equation.

The authors build upon their previous work [11] and the broader literature [1, 2, 3, 4, 5, 8, 10, 15, 18, 21] to establish conditions under which such pairs exist.

2. Main Conjectures and Results

Conjecture 1 (Sufficient Conditions)

The authors propose that if $a, b, c$ are nonzero integers and at least one of $ac$, $bc$, or $(a+b)c$ is a perfect square, then the equation $ax^2 + by^2 = cz^2$ is partition regular with respect to $x, y$ .

Theorem 1.2 (Unconditional Result)

If $ac$ or $bc$ is a perfect square, the equation is partition regular with respect to $x, y$ .

This extends previous results covering cases where multiple quantities are squares.
It resolves specific open problems, such as the partition regularity of $x^2 + 2y^2 = z^2$ with respect to $x, y$ and $y, z$ .

Theorem 1.3 (Conditional Result)

Assuming Conjecture 2 (an Elliott-type conjecture regarding the vanishing of correlations of aperiodic multiplicative functions along irreducible binary quadratic forms), if $(a+b)c$ is a perfect square, the equation is partition regular with respect to $x, y$ .

Conjecture 2: If $P_1, P_2$ are irreducible binary quadratic forms that are not multiples of each other, they are "good for vanishing of correlations" of aperiodic multiplicative functions.
This result, combined with Theorem 1.2, implies that Conjecture 2 implies Conjecture 1.

Theorem 1.4 (Unconditional Density Result)

Even without Conjecture 2, if $(a+b)c$ is a square, the authors prove a density result for pretentious multiplicative functions. Specifically, for any finite collection of pretentious multiplicative functions $f_j$ , there exist distinct $x, y$ in a set of positive multiplicative density such that $f_j(x)$ and $f_j(y)$ lie in a specific arc near 1, satisfying the equation.

Necessity and Limitations

The paper investigates whether the conditions in Conjecture 1 are necessary.

Proposition 1.15: If the equation is partition regular with respect to $x, y$ , then at least one of $ac, bc, (a+b)c$, or $ab(a+b)c$ must be a perfect square.
Propositions 1.16 & 1.17: The condition that $ab(a+b)c$ is a square is not sufficient. The authors provide counterexamples (e.g., $x^2 + 2y^2 = 6z^2$ and $3x^2 + 5y^2 = 30z^2$ ) where the product is a square, yet the equation fails to be partition regular due to modular obstructions (mod 4 and mod 8).

3. Methodology and Technical Framework

The proofs rely on a combination of ergodic theory, analytic number theory, and the theory of multiplicative functions. The strategy follows a "circle method" approach over the group of multiplicative functions $\mathcal{M}$ .

3.1. Density Regularity

The authors first establish density regularity (existence of solutions in sets of positive multiplicative density), which implies partition regularity. They utilize multiplicative Følner sequences $\Phi = (\Phi_K)$ to define upper multiplicative density.

3.2. Parametric Reformulation

The problem is reduced to studying the density regularity of pairs of polynomials $(P_1, P_2)$ derived from the parametric solutions of the quadratic equation.

Case 1 (Theorem 1.12): $P_1(m,n) = \alpha m^2 + \beta n^2$ and $P_2(m,n) = \gamma mn$ .
Case 2 (Theorem 1.13): $P_1, P_2$ are distinct, irreducible binary quadratic forms with equal leading coefficients.

3.3. Decomposition of Multiplicative Functions

The core analytical tool is the decomposition of multiplicative functions into:

Pretentious functions: Those that correlate with Dirichlet characters twisted by powers ( $f \sim \chi \cdot n^{it}$ ).
Aperiodic functions: Those that do not correlate with any such structure.

The proof strategy involves showing that:

Aperiodic contributions (Minor Arcs): Correlations vanish. This is achieved by extending the Daboussi-Kátai orthogonality criterion to quadratic number fields (Theorem 3.1). A key difficulty addressed is the presence of infinitely many units in real quadratic fields (positive discriminant), resolved using recent results by Sun [24] on "regular" elements.
Pretentious contributions (Major Arcs): Concentration estimates are used to show that the averages are positive. This relies on new concentration estimates for multiplicative functions along arbitrary binary quadratic forms (Propositions 4.8 and 4.9).

3.4. Key Technical Innovations

Concentration Estimates: The paper proves that for pretentious functions, the values of $f(P(Qm+a, Qn+b))$ concentrate around a specific value determined by the character and the twist. This is non-trivial for irreducible quadratic forms, especially those with positive discriminant.
Uniformity: The authors establish uniformity in the parameters of the concentration estimates, which is crucial for handling the "major arcs" integration.
Weighted Averages: The use of specific weights $w_{\delta, c, P_1, P_2}$ allows the authors to isolate regions where the ratio of the polynomials is close to a constant, ensuring the positivity of the integral.

4. Significance and Claims

The paper claims to make significant progress on the partition regularity of generalized Pythagorean pairs, specifically:

Verification of Partial Cases: It unconditionally verifies Conjecture 1 for cases where $ac$ or $bc$ is a square.
Conditional Verification: It reduces the remaining cases to a natural conjecture (Conjecture 2) regarding the correlation of multiplicative functions, a problem of independent interest in analytic number theory.
Methodological Extension: It extends the authors' previous methods (from [11]) to handle more general quadratic forms, specifically overcoming the difficulties posed by positive discriminants (real quadratic fields) and pairs of distinct irreducible forms.
Necessity Analysis: It clarifies the boundaries of the conjecture by proving that certain algebraic conditions are necessary but not sufficient, highlighting the role of modular arithmetic obstructions.

The authors explicitly state that their methods do not yet establish full partition regularity (all three variables) for the general equation, nor do they resolve the open cases where the necessary conditions are met but the sufficient conditions of the conjecture are not obviously satisfied (e.g., $x^2 + 17y^2 = 34z^2$ ). The results are presented as a substantial step forward, bridging the gap between known results for specific equations and the general theory of quadratic partition regularity.