Abstract:We combine two of Igusa's conjectures with recent semi-continuity results by Mustaţă and Popa to form a new, natural conjecture about bounds for exponential sums. These bounds have a deceivingly simple and general formulation in terms of degrees and dimensions only. We provide evidence consisting partly of adaptations of already known results about Igusa's conjecture on exponential sums, but also some new evidence like for all polynomials in up to $4$ variables. We show that, in turn, these bounds imply consequences for Igusa's (strong) monodromy conjecture. The bounds are related to estimates for major arcs appearing in the circle method for local-global principles.
What problem does this paper attempt to address?
The core problem that this paper attempts to solve is about the estimation of exponential sums and its connection with the monodromy conjecture. Specifically, the author combines Igusa's two conjectures regarding exponential sums and monodromy, and uses the recent semi - continuity results of Mustaţă and Popa on the minimal exponent to propose a new natural conjecture, namely, the bound on exponential sums.
### Problem Background
The main object discussed in the paper is the finite exponential sum:
\[
E_f(N, \xi) := \left| \frac{1}{N^n} \sum_{x \in (\mathbb{Z}/N\mathbb{Z})^n} \xi^{f(x)} \right|
\]
where \( f \) is a polynomial with integer coefficients defined on \( n \) variables, \( N \) is a positive integer, and \( \xi \) is a primitive \( N \)-th root of unity in the complex number field. The key parameters here include:
- \( n \): the number of variables of the polynomial
- \( d \): the degree of the polynomial
- \( s \): the complex affine dimension of the critical point set of the homogeneous part \( f_d \) of \( f \)
### Main Objective
The goal of the paper is to propose and verify the following conjecture:
**Conjecture 1**: For given \( f, n, s, d \) and any \( \epsilon>0 \), there exists a constant \( c = c(f,\epsilon) \) such that for all positive integers \( N \) and all primitive \( N \)-th roots of unity \( \xi \), we have:
\[
E_f(N, \xi) \ll N^{-\frac{n - s}{d}+\epsilon}
\]
This conjecture not only simplifies Igusa's original problem regarding exponential sums, but also provides a new approach for solving more general cases of irrational singularities.
### Connection with the Monodromy Conjecture
Furthermore, the paper also explores the relationship between this conjecture and Igusa's strong monodromy conjecture. Specifically, Conjecture 1 implies the strong monodromy conjecture within certain ranges. For example, Conjecture 1 implies that when the poles of the local zeta function have real parts strictly between \( -\frac{n - s}{d} \) and 0, the only pole is \( - 1\).
### Evidence Provided
To support Conjecture 1, the author provides various evidence, including but not limited to:
1. **Adaptation of Known Results**: such as Birch's results and the conjecture of Denef and Sperber.
2. **New Evidence**: such as proving for polynomials with at most 4 variables.
3. **Verification under Specific Conditions**: such as cube - free and \( (d + 2)\)-th power - free \( N \).
These evidences show that although the conjecture seems simple, it is quite difficult to prove it in general.
### Summary
In summary, this paper aims to propose a new conjecture about exponential sums by combining Igusa's two conjectures and the latest semi - continuity results, and to explore its relationship with the monodromy conjecture. This not only provides a new perspective for the study of exponential sums, but also lays the foundation for solving more general mathematical problems.