A Stothers-Mason theorem with a difference radical (Q2031732)

From MaRDI portal
Revision as of 21:40, 19 March 2024 by Openalex240319060354 (talk | contribs) (Set OpenAlex properties.)
scientific article
Language Label Description Also known as
English
A Stothers-Mason theorem with a difference radical
scientific article

    Statements

    A Stothers-Mason theorem with a difference radical (English)
    0 references
    0 references
    10 June 2021
    0 references
    Let \(P\) be a polynomial in \(\mathbb{C}[z]\). The radical of \(P\), denoted by rad(P), is defined to be the product of distinct linear factors of \(P\). The Stothers-Mason theorem states that if \(a, b, c\) are three relatively prime polynomials, not all identically zero, satisfy \(a + b = c\), then \(\deg (c) \leq \mathrm{deg}( \mathrm{rad} (abc)) {- 1}\). The Stothers-Mason theorem is generalized in many directions, i.e., to sums in one-dimensional functions field, to sums of pairwise relatively prime polynomials of several variables etc. Let \(p \not\equiv 0\) be a polynomial in \(\mathbb{C}[z]\) and \(\kappa \in \mathbb{C}\backslash \{0\}\). Then the \(\kappa\)-difference radical of \(p\) is defined by the authors as \(\widetilde{\mathrm{rad}}_{\kappa}(p) = \prod\limits_{w \in \mathbb{C}}(z - w)^{d_{\kappa}(w)}\), where \(d_{\kappa}(w) = \text{ord}_{w}(p) - \min \{\text{ord}_{w}(p), \text{ord}_{w + \kappa}(p)\}\), \(\text{ord}_{w}(p) \geq 0\) being the order of zero of the polynomial \(p\) at \(w \in \mathbb{C}\). The authors prove a difference analogue of the Stothers-Mason theorem as follows: If \(a, b, c\) are relatively prime polynomials in \(\mathbb{C}[z]\), not all constants, such that \(a + b = c,\) then \(\max \{\deg (a), \deg (b), \deg (c)\} \leq \deg\widetilde{\mathrm{rad}}_{\kappa}(p)(a) + \deg \widetilde{\mathrm{rad}}_{\kappa}(p) d_{\kappa}(b) + \deg \widetilde{\mathrm{rad}}_{\kappa}(p) d_{\kappa}(c) - 1\), where \(\kappa \in \mathbb{C}\backslash \{0\}\). The authors also extend this result to \(m + 1\) polynomials. In order to study the polynomial solutions of the super-Fermat functional equation, the authors introduce the factorial of a polynomial \(p\) in \(\mathbb{C}[z]\) as \([p]^{\bar n}_{\kappa} = p(z)p(z + \kappa)\cdots p(z + (n - 1)\kappa)\), where the shift \(\kappa \in \mathbb{C}\backslash \{0\}\). The authors also prove that if \(\kappa \in \mathbb{C}\backslash \{0\}\), \(a, b, c\) are polynomials in \(\mathbb{C}[z]\), not all constants, \([a]^{\bar n}_{\kappa}\), \([b]^{\bar n}_{\kappa}\) and \([c]^{\bar n}_{\kappa}\) are relatively prime and satisfy \([a]^{\bar n}_{\kappa} + [b]^{\bar n}_{\kappa} = [c]^{\bar n}_{\kappa},\) then \(n \leq 2\). If at least one of \(a, b, c\) is a constant, then \(n = 1\). They also extend the result to \(m + 1\) (\(m \geq 2\)) polynomials. In the last section of the paper the authors consider non-polynomial entire solutions of the difference super-Fermat functional equation.
    0 references
    Stothers-Mason theorem
    0 references
    difference radical
    0 references
    polynomial
    0 references
    super-Fermat functional equation
    0 references

    Identifiers