Mixed volume computation via linear programming (Q5928410)

The mixed volume of the support \(\mathcal A = (\mathcal A_1, \ldots, \mathcal A_n)\) of a polynomial system \(P(x)=(p_1(x),\dots\), \(p_n(x))\) is important for the approximation of the isolated zeros of \(P(x)\). It can be computed from the fine mixed cells of a subdivision of \(\mathcal A\). This subdivison can be obtained via lifting functions and lower facets. The authors present a new algorithm where the lower facets are computed using a tailored version of the simplex algorithm for linear programming. They show the superiority of their methods over others by numerical results.