Abstract:The inhomogeneous Khintchine-Groshev Theorem is a classical generalization of Khintchine's Theorem in Diophantine approximation, by approximating points in $\mathbb{R}^m$ by systems of linear forms in $n$ variables. Analogous to the question considered by Duffin and Schaeffer for Khintchine's Theorem (which is the case $m = n = 1$), the question arises for which $m,n$ the monotonicity can be safely removed. If $m = n = 1$, it is known that monotonicity is needed. Recently, Allen and Ramirez showed that for $mn \geq 3$, the monotonicity assumption is unnecessary, conjecturing this to also hold when $mn = 2$. In this article, we confirm this conjecture for the case $(m,n)=(1,2)$ whenever the inhomogeneous parameter is a non-Liouville irrational number. Furthermore, under mild assumptions on the approximation function, we show an asymptotic formula (with almost square-root cancellation), which is not even known for homogeneous approximation. The proof makes use of refined overlap estimates in the 1-dimensional setting, which may have other applications including the inhomogeneous Duffin-Schaeffer conjecture.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is whether the monotonicity assumption in the inhomogeneous Khintchine - Groshev theorem can be removed, especially in the case of \((m, n)=(1, 2)\). Specifically, the author confirms the conjecture of Allen and Ramírez, that is, for inhomogeneous parameters of non - Liouville irrational numbers, when \((m, n)=(1, 2)\), the monotonicity assumption can be removed.
### Main problem analysis
#### 1. **Inhomogeneous Khintchine - Groshev theorem**
- **Background**: The classical Khintchine - Groshev theorem is a generalization of the linear form system in Diophantine approximation. It allows approximation in high - dimensional space and introduces inhomogeneous parameters.
- **Core of the problem**: For which \(m\) and \(n\) can the monotonicity assumption be removed in the inhomogeneous Khintchine - Groshev theorem?
#### 2. **Existing results**
- When \(m = n=1\), the monotonicity assumption is necessary (proved by Duffin and Schaeffer).
- Recent research shows that when \(mn\geq3\), the monotonicity assumption can be removed (the work of Allen and Ramírez).
- For the case of \(mn = 2\), especially \((m, n)=(1, 2)\) or \((m, n)=(2, 1)\), whether the monotonicity assumption can be removed is still an open problem.
#### 3. **Contributions of this paper**
- The author proves that when \((m, n)=(1, 2)\) and the inhomogeneous parameter is a non - Liouville irrational number, the monotonicity assumption can be removed.
- In addition, the author also shows that in this case, an asymptotic formula (with an error term of almost square - root level) can be obtained, which is a result not yet known in uniform approximation.
### Formula summary
- **Lebesgue measure formula**:
\[
\lambda_{nm}(A_{\psi,\gamma}^{n,m})=\begin{cases}
0 & \text{if }\sum_{q\in\mathbb{N}}q^{n - 1}\psi(q)^{m}<\infty\\
1 & \text{if }\sum_{q\in\mathbb{N}}q^{n - 1}\psi(q)^{m}=\infty
\end{cases}
\]
- **Asymptotic formula**:
\[
N(\alpha,Q,\gamma)=\Psi(Q)+O\left(\Psi(Q)^{1/2}\log(\Psi(Q)^{3/2+\delta})\right),\quad Q\rightarrow\infty
\]
where \(\Psi(Q) = 16\sum_{q\leq Q}q\psi(q)+8\sum_{q\leq Q}\psi(q)\)
### Conclusion
Through these results, the author not only confirms the conjecture of Allen and Ramírez, but also provides a more accurate asymptotic formula under specific conditions, which has not been achieved in uniform approximation. This result provides new insights into the theory of inhomogeneous Diophantine approximation and lays the foundation for further research.