On the bifurcation set of a polynomial function and Newton boundary. II (Q1816453)

Let \(f:\mathbb{C}^n\to \mathbb{C}\) be a polynomial function and \(\text{supp}(f)= \{\nu\in\mathbb{N}^n:a_\nu \neq 0\}\). For \(\Delta\in{\mathcal B}\) (the set of bad faces of \(\overline{\text{supp}(f)}\)) \[ \Sigma_\Delta=\{f_\Delta(z^0):z^0\in (\mathbb{C}\setminus \{0\})^n,\text{grad } f_\Delta(z^0)=0\}\quad (f(z)=\sum_{\nu\in \mathbb{N}^n}a_\nu z^\nu). \] The main result is the following: let \(f\) be a polynomial Newton nondegenerate, not convenient and \(f(0)=0\) and let \(K\) be a conic polyhedron constructed in a suitable way. For a bad face \(\Delta\in {\mathcal B}\) let \(t_0\in\Sigma_\Delta\) be such that \(t_0\notin \{0\}\cup\Sigma_f\) and let a hypothesis for \(t_0\) be satisfied. Then \(t_0\in B_f\) and for \(t\notin B_f\) is valid \[ \chi(f^{-1}(t_0))-\chi(f^{-1}(t))=(-1)^{n-1}\sum_{p^0\in X_{t_0}\cap{\mathcal N}_D}[\mu(X_{t_0},p^0)+\sum_{\Phi[\sigma]\subseteqq{\mathcal N}_D,p^0\in\overline{\Phi[\sigma]}} \mu(X_{t_0}\cap \overline{\Phi[\sigma]},p^0)] \] where \(\mu(X_{t_0},p^0)\), \(\mu(X_{t_0}\cap\overline{\Phi[\sigma]},p^0)\) are Milnor numbers in a point \(p^0\in X_{t_0}\cap{\mathcal N}_D\) such that \(p^0\in\overline{\Phi[\sigma]}\). [For part I see \textit{A. Nemethi} and the author, Publ. Res. Inst. Math. Sci. 26, No. 4, 681-689 (1990; Zbl 0736.32024)].