Paucity phenomena for polynomial products

Victor Y. Wang,Max Wenqiang Xu
DOI: https://doi.org/10.1112/blms.13095
2024-06-06
Abstract:Let $P(x)\in \mathbb{Z}[x]$ be a polynomial with at least two distinct complex roots. We prove that the number of solutions $(x_1, \dots, x_k, y_1, \dots, y_k)\in [N]^{2k}$ to the equation \[ \prod_{1\le i \le k} P(x_i) = \prod_{1\le j \le k} P(y_j)\neq 0 \] (for any $k\ge 1$) is asymptotically $k!N^{k}$ as $N\to +\infty$. This solves a question first proposed and studied by Najnudel. The result can also be interpreted as saying that all even moments of random partial sums $\frac{1}{\sqrt{N}}\sum_{n\le N}f(P(n))$ match standard complex Gaussian moments as $N\to +\infty$, where $f$ is the Steinhaus random multiplicative function.
Number Theory,Probability
What problem does this paper attempt to address?
The problem that this paper attempts to solve is related to the paucity phenomena of polynomial products. Specifically, the author studied Diophantine equations of the following form: \[ \prod_{1 \leq i \leq k} P(x_i) = \prod_{1 \leq j \leq k} P(y_j) \neq 0 \] where \(P(x)\in\mathbb{Z}[x]\) is a polynomial with integer coefficients having at least two distinct complex roots. For given integers \(k\geq1\) and \(N\geq1\), the author defined \(A_{P,2k}([N])\) to represent the number of solutions \((x_1,x_2,\ldots,x_k,y_1,y_2,\ldots,y_k)\in[N]^{2k}\) that satisfy the above equation. ### Core of the research problem The author's goal is to prove that as \(N\rightarrow+\infty\), the number of non - trivial solutions is very small compared to the number of all solutions. More specifically, they attempt to prove that as \(N\rightarrow+\infty\), almost all solutions are trivial solutions, that is: \[ \lim_{N\rightarrow+\infty}\left(\frac{A_{P,2k}([N])}{N^k}\right)=k! \] This problem was first proposed and studied by Najnudel. By solving this problem, the author not only answered an important mathematical question, but also provided a theoretical basis for the matching of high - order moments of random multiplicative functions and standard complex Gaussian moments. ### Main results The author's main result is to prove that for polynomials \(P(x)\) satisfying certain necessary conditions, the above - mentioned paucity phenomena do exist. Specifically, let the degree \(d\geq2\) of \(P(x)\in\mathbb{Z}[x]\) and \(P(x)\) has at least two distinct complex roots, then: \[ A_{P,2k}([N])=k!N^k+O_{P,k,\varepsilon}(N^{k-\frac{1}{6e_P}+\varepsilon}) \] Here \(e_P\) is the maximum multiplicity of any complex root of \(P(x)\), \(1\leq e_P\leq d - 1\). ### Applications This result can be interpreted as: for the Steinhaus random multiplicative function \(f\), the even moments of its partial sums match the moments of the standard complex Gaussian distribution as \(N\rightarrow+\infty\). This provides an important basis for further research on the central limit theorem of random multiplicative functions. In conclusion, this paper solved the key problem regarding the paucity of polynomial products and provided new insights into the study of random multiplicative functions.