On the moduli space of quasi-homogeneous functions (Q2160332)

Let \(f(x,y)\in \mathbb{C}\{x,y\}\) be a square-free quasi-homogeneous (but not homogeneous) function germ. Essentially the following types are interesting: \[ \text{II}:\quad f(x,y)=x\cdot\prod^n_{j=1}(y-\lambda_j x^q),\quad q>1, \] \[ \text{III}:\quad f(x,y)=xy\cdot\prod^n_{j=1}(y^p-\lambda_j x^q),\quad q>p. \] One compares the analytic classification of such function-germs to their bi-Lipschitz classification. The analytic classification gives the moduli spaces, \(\mathbb{Z}_2\times\frac{\mathrm{Symm}(\mathbb{C}_\Delta(n))}{\mathrm{Aff}(\mathbb{C})}\) for type II, and \(\mathbb{Z}_2\times\mathbb{Z}_2\times\frac{\mathrm{Symm}(\mathbb{C}^*_\Delta(n))}{GL(1,\mathbb{C})}\) for type III. The bi-Lipschitz equivalence class is encoded (in this case) by the Henry-Parusiński invariant. The authors show: this invariant determines the analytic equivalence class up to a finite ambiguity. More precisely: Theorems 2 and 3. (Type II.) To each value of the Henry-Parusiński invariant correspond at most \(n^{n-3}\) analytic types. (Type III.) To each value of the Henry-Parusiński invariant correspond at most \(n^{n-2}\) analytic types. In both cases the equalities hold generically. As the corollary the authors obtain: Theorem 4. Let \(f_t\) be a continuous family of quasi-homogeneous (not homogeneous) function-germs with isolated singularity. If the Henry-Parusiński invariant is constant in this family then the family is analytically trivial.