Abstract:Recently, Maesaka, Watanabe, and the third author discovered a phenomenon where the iterated integral expressions of multiple zeta values become discretized. In this paper, we extend their result to the case of multiple polylogarithms and provide two proofs. The first proof uses the method of connected sums, while the second employs induction based on the difference equations that discrete multiple polylogarithms satisfy. We also investigate several applications of our main result.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to generalize the discretization phenomenon of the iterative integral expressions of multiple polylogarithms (MPLs) from multiple zeta values (MZVs) to a broader situation. Specifically, the authors hope to explore whether the discretization phenomenon of "series = integral" - type expressions is also applicable in multiple polylogarithm functions and provide strict mathematical proofs.
### Background and Motivation
1. **Discretization of Multiple Zeta Values**:
- Recently, Maesaka, Watanabe and the third author discovered a phenomenon that the iterative integral expressions of multiple zeta values can be discretized. This means that without taking the limit, the integral expressions can still be valid in discrete form.
2. **Generalization of Multiple Polylogarithms**:
- The goal of this paper is to generalize this discretization phenomenon to multiple polylogarithm functions. Multiple polylogarithm functions are a broader class of special functions, which have important applications in mathematical physics, number theory and other fields.
### Main Results
- **Theorem 1.2**: The authors proved the discretization formula of the iterative integral expressions of multiple polylogarithm functions by two methods. The first method uses the technique of connected sums, and the second method is based on the inductive proof using the difference equations satisfied by the discrete multiple polylogarithm functions.
- **Specific Form of Discretization**:
\[
\sum_{0 < n_1 < \cdots < n_r < N} \frac{1}{n_1^{k_1} \cdots n_r^{k_r}} \prod_{i = 1}^r \left( \frac{Nx_i + 1}{n_i} \right)=(-1)^r \sum_{\substack{1 \leq n_{j,1} \leq \cdots \leq n_{j,k_j} < N \\ (1 \leq j \leq r)}} \prod_{j = 1}^r \frac{1}{(n_{j,1} - Nx_j)n_{j,2} \cdots n_{j,k_j}}
\]
### Applications and Significance
- **Theoretical Significance**: This result not only provides a new perspective on the properties of multiple polylogarithm functions, but also provides a methodological reference for studying the discretization phenomena of other special functions.
- **Practical Applications**: In numerical calculations and algorithm design, the discretization formula can be used to approximately calculate complex integral expressions, thereby improving computational efficiency.
In conclusion, through rigorous mathematical derivations, this paper successfully generalizes the discretization phenomenon of multiple zeta values to multiple polylogarithm functions and provides detailed proofs and application examples for this.