Abstract:We prove uniform upper bounds on the number of integral points of bounded height on affine varieties. If $X$ is an irreducible affine variety of degree $d\geq 4$ in $\mathbb{A}^n$ which is not the preimage of a curve under a linear map $\mathbb{A}^n\to \mathbb{A}^{n-\dim X+1}$, then we prove that $X$ has at most $O_{d,n,\varepsilon}(B^{\dim X - 1 + \varepsilon})$ integral points up to height $B$. This is a strong analogue of dimension growth for projective varieties, and improves upon a theorem due to Pila, and a theorem due to Browning-Heath-Brown-Salberger.
Our techniques follow the $p$-adic determinant method, in the spirit of Heath-Brown, but with improvements due to Salberger, Walsh, and Castryck-Cluckers-Dittmann-Nguyen. The main difficulty is to count integral points on lines on an affine surface in $\mathbb{A}^3$, for which we develop point-counting results for curves in $\mathbb{P}^1\times \mathbb{P}^1$.
We also formulate and prove analogous results over global fields, following work by Paredes-Sasyk.
What problem does this paper attempt to address?
### What problem does this paper attempt to solve?
This paper aims to solve the problem of a uniform upper bound on the number of integral points on affine varieties. Specifically, the author attempts to prove a strong version of the dimension - growth theorem for the number of integral points of a specific type of irreducible affine variety \(X\) under bounded height.
#### Main problem description:
1. **Upper bound estimation of the number of integral points**:
- If \(X\) is an irreducible affine variety with dimension \(m\), degree \(d\geq4\), and is not the pre - image of some curve under some linear mapping, then the author proves that the number of integral points of \(X\) with height not exceeding \(B\) is at most \(O_{d,n,\varepsilon}(B^{\dim X - 1+\varepsilon})\).
- This result is an improvement of the theorems of Pila and Browning - Heath - Brown - Salberger.
2. **Improved methods and techniques**:
- The author uses the \(p\)-adic determinant method, a technique proposed by Heath - Brown, combined with improvements by Salberger, Walsh, and Castryck - Cluckers - Dittmann - Nguyen.
- The key difficulty lies in calculating the number of integral points on curves in three - dimensional affine space. To this end, the author develops point - counting results for curves on \(\mathbb{P}^1\times\mathbb{P}^1\).
3. **Generalization to global fields**:
- The author also formulates and proves similar results on global fields, which extends previous work and is applicable to the case of positive characteristic.
#### Mathematical expressions:
- Let \(X\subset\mathbb{A}^n\) be an irreducible affine variety with dimension \(m\), degree \(d\geq4\), and not the pre - image of some curve under some linear mapping. Then:
\[
N_{\text{aff}}(X, B)\ll_{d,n,\varepsilon}B^{\dim X - 1+\varepsilon}
\]
where \(N_{\text{aff}}(X, B)\) represents the number of integral points with height not exceeding \(B\).
- For a surface \(X\subset\mathbb{A}^3\) in three - dimensional affine space, if \(X\) is irreducible and controlled, then:
\[
N_{\text{aff}}(X, B)\ll_{d,n,\varepsilon}B^{m - 1+\varepsilon},\quad\text{when}\,d\geq4
\]
\[
N_{\text{aff}}(X, B)\ll_{n,\varepsilon}B^{m - 2+\frac{2}{\sqrt{3}}+\varepsilon},\quad\text{when}\,d = 3
\]
#### Conclusion:
Through these results, the author not only improves existing theorems but also proposes a stronger dimension - growth conjecture in higher dimensions and provides new tools and techniques for future research.