Schinzel's multivariate pullback factorization conjecture

Let HC[y±1]H\in\mathbb{C}[{\boldsymbol{y}}^{\pm1}] and let k2k\geq 2. For AZn×kA\in\mathbb{Z}^{n\times k}, write H(tA)H({\boldsymbol{t}}^A) for the pullback of HH under the monomial map ttA{\boldsymbol{t}}\mapsto{\boldsymbol{t}}^A, and let SS be the multiplicative subset generated by all f(td)f({\boldsymbol{t}}^{\boldsymbol{d}}) with fC[z±1]f\in\mathbb{C}[z^{\pm1}] and dZk{\boldsymbol{d}}\in\mathbb{Z}^k.

Multivariate pullback factorization conjecture. There is a finite set of matrices ΩZn×n\Omega\subset\mathbb{Z}^{n\times n} such that, for each AZn×kA\in\mathbb{Z}^{n\times k}, there are NΩN\in\Omega and BZn×kB\in\mathbb{Z}^{n\times k} with A=NBA=NB such that if PP is an irreducible factor of H(yN)H({\boldsymbol{y}}^N), then P(tB)P({\boldsymbol{t}}^B) is, as an element of C[t±1]S\mathbb{C}[{\boldsymbol{t}}^{\pm1}]_S, either a unit or an irreducible factor of H(tA)H({\boldsymbol{t}}^A).

This conjecture would give a finite description of irreducible factors that genuinely depend on more than one variable; the remaining univariate factors become units after localization. Its status is not resolved in the supplied text.

Sources & referencesView supporting material

Primary source

Francesco Amoroso and Martín Sombra, “Factorization of bivariate sparse polynomials”, arXiv:1710.11479 (2018).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.