Theresa C. Anderson,Ayla Gafni,Kevin Hughes,Robert J. Lemke Oliver,David Lowry-Duda,Frank Thorne,Jiuya Wang,Ruixiang Zhang
Abstract:We study the number of degree $n$ number fields with discriminant bounded by $X$. In this article, we improve an upper bound due to Schmidt on the number of such fields that was previously the best known upper bound for $6 \leq n \leq 94$.
What problem does this paper attempt to address?
### What problem does this paper attempt to solve?
This paper aims to improve the upper bound estimates of the number of number fields with small degrees. Specifically, the authors studied the number of number fields of degree \(n\) with bounded discriminants and improved Schmidt's previously best - known upper bounds for \(6\leq n\leq94\).
#### Background and Main Contributions of the Paper
1. **Problem Definition**:
- Define \(N_n(X):=\#\{K / \mathbb{Q}:[K:\mathbb{Q}]=n,|Disc(K)|\leq X\}\), that is, calculate the number of number fields of degree \(n\) over the rational number field \(\mathbb{Q}\) whose discriminant absolute values do not exceed \(X\).
2. **Existing Results**:
- According to Hermite's classical theorem, \(N_n(X)\) is finite.
- Schmidt [Sch95] proposed an optimized version of Hermite's argument, showing that \(N_n(X)\ll_n X^{\frac{n + 2}{4}}\).
- It is conjectured that the actual optimal upper bound is \(O_n(X)\).
3. **Improvement Goals**:
- For \(n\leq5\), there already exist exact asymptotic formulas.
- For larger \(n\), the works of Ellenberg and Venkatesh [EV06], Couveignes [Cou20], and Lemke Oliver and Thorne [LOT20] et al. have significantly improved Schmidt's upper bounds.
- In particular, for \(n\geq95\), [LOT20] gives the currently best upper bound \(N_n(X)\ll_n X^{c(\log n)^2}\), where \(c\leq1.564\).
- But in the range of \(6\leq n\leq94\), Schmidt's upper bound has still been the best result in the past 25 years.
4. **The Main Theorem of This Paper**:
- The authors proposed a new upper bound: for each \(n\geq6\) and any \(\epsilon>0\), we have
\[
N_n(X)\ll_{n,\epsilon}X^{\frac{n + 2}{4}-\frac{1}{4n - 4}+\epsilon}.
\]
- Although this improvement may not be optimal, it achieves significant progress through different methods.
5. **Method Overview**:
- The authors reconstructed Schmidt's method and were not limited to polynomials with zero traces.
- Count polynomials with "particularly small" discriminants and "particularly square - power" discriminants separately.
- Use non - Archimedean Fourier transform techniques to study the set of polynomials whose discriminants are divisible by large prime - power numbers.
- Combine local density estimates and analytical techniques in different frameworks to improve the polynomial counting method.
Through these improvements, the authors successfully provided better upper - bound estimates than Schmidt in the range of \(6\leq n\leq94\).