Abstract:We provide a new proof of the multivariate version of Christol's theorem about algebraic power series with coefficients in finite fields, as well as of its extension to perfect ground fields of positive characteristic obtained independently by Denef and Lipshitz, Sharif and Woodcok, and Harase. Our proof is elementary, effective, and allows for much sharper estimates. We discuss various applications of such estimates, in particular to a problem raised by Deligne concerning the algebraicity degree of reductions modulo $p$ of diagonals of multivariate algebraic power series with integer coefficients.
What problem does this paper attempt to address?
The core problem that this paper attempts to solve is to provide a new proof of the multivariate version of Christol's theorem and extend it to perfect fields of positive characteristic. In addition, the author also seeks more precise estimation methods to improve the understanding and application of algebraic power series.
Specifically, the main contributions of the paper include:
1. **New proof and improved estimation**: The author provides a new, elementary proof method for the multivariate version of Christol's theorem and its extension over perfect fields of positive characteristic. This new method not only simplifies the proof process but also allows for more precise estimation results.
2. **Scope of application**: The author discusses various applications of these estimates, in particular, solving a problem proposed by Deligne, namely, the degree of algebraicity of the diagonal of a multivariate algebraic power series with integer coefficients modulo \( p \).
3. **Algorithmic consequences**: The paper also explores the algorithmic consequences of the main result (Theorem 1.2), in particular, when calculating the \( M \)-th coefficient of the diagonal of an algebraic power series, the number of operations in \( F_p \) is of the logarithmic level of \( M \) and approximately linear level of \( p^{n + 1} \).
### Formula representation
To ensure the correctness and readability of the formulas, the following are some key formulas in the paper represented in Markdown format:
- **Frobenius mapping**:
Let \( k \) be a perfect field with characteristic \( p>0 \), then the Frobenius endomorphism \( F \) maps \( x \) to \( x^p \), that is,
\[
F(x)=x^p
\]
- **Power series expansion**:
For each \( r\in\{0,\ldots,p - 1\}^n \), the section operator \( S_r \) is defined as
\[
S_r(f):=F^{-1}(f_r)
\]
where \( f_r \) is the coefficient of \( f \) at \( t^r \).
- **Newton polytope**:
Let \( A(t,y) \) be a multivariate polynomial, its Newton polytope \( NP(A) \) is defined as
\[
NP(A):=\text{convex hull in }\mathbb{R}^{n + 1}\text{ of the tuples }(i,j)\text{ with }a_{i,j}\neq0
\]
Through these improvements, the paper significantly improves the understanding of multivariate algebraic power series and provides more effective tools to deal with related problems.