Abstract:We prove significant power savings for the error term when counting abelian extensions of number fields (as well as the twisted version of these results for nontrivial Galois modules). In some cases over $\mathbb{Q}$, these results reveal lower order terms following the same structure as the main term that were not previously known. Assuming the generalized Lindelöf hypothesis for Hecke $L$-functions, we prove square root power savings for the error compared to the order of the main term.
What problem does this paper attempt to address?
This paper attempts to achieve a significant improvement of the error term in the problem of counting Abelian extensions in number theory. Specifically, the author aims to prove that when calculating the Abelian extensions of a number field \(K\) (and the twisted versions of these results), the error term can be significantly reduced. The following are the main objectives and contributions of this paper:
1. **Significantly reducing the error term**: The author proves that when calculating the number of Abelian extensions, the error term can be significantly reduced. In particular, in some cases, these results reveal previously unknown lower - order terms, which have the same structure as the main term.
2. **Assuming the generalized Lindelöf conjecture**: Under the assumption of the generalized Lindelöf conjecture, the author proves a square - root - level saving of the error term compared to the main term.
3. **Introducing a new invariant**: In order to capture the power - saving of the error term, the author defines a new invariant \(\theta(K, G)\) and proves its upper bound. This helps to describe the size of the error term more precisely.
4. **Existence of lower - order terms**: The author also studies the existence of lower - order terms and proves that some lower - order terms do exist in specific cases. For example, for certain Abelian groups \(G\), the generating series has poles at specific points, and these poles correspond to lower - order terms.
5. **Generalization to twisted Abelian extensions**: In addition to standard Abelian extensions, the author also studies the case of twisted Abelian extensions and proves similar power - saving results.
### Formula summary
- **Asymptotic formula for the main term**:
\[
\# \text{Surj}(G_K, G; X) \sim c X^{\frac{1}{a(G)}} (\log X)^{b(K, G)- 1}
\]
where \(a(G)=\min_{g \neq 1} \text{ind}(g)\), and \(b(K, G)\) is the number of orbits of non - unit conjugacy classes under the minimum index.
- **Power - saving of the error term**:
\[
\# \text{Surj}(G_K, G; X)=X^{\frac{1}{a(G)}} P(\log X)+O(X^{\frac{1}{a(G)}-\delta})
\]
where \(\delta > 0\) is a positive number representing the power - saving.
- **New invariant \(\theta(K, G)\)**:
\[
\theta(K, G)=\inf \left\{\theta : D_{K, G}(s) \text{ has a meromorphic continuation on } \Re(s)>\theta\right\}
\]
Through these results, the author not only improves the understanding of the problem of counting Abelian extensions but also provides new tools and methods for future research.