Associated forms in classical invariant theory and their applications to hypersurface singularities (Q476767): Difference between revisions

Let \(Q^m_n\) be the vector space of forms of degree \(m\) on \(\mathbb{C}^n\). Let \(f\in Q^m_n\) with non--zero discriminant and \(M(f)\) the Milnor algebra of the isolated singularity at the origin of the hypersurface in \(\mathbb{C}^n\) defined by \(f\). The algebra \(M(f)\) gives rise to a form on \(\mathfrak{m}/\mathfrak{m}^2\) with values in the socle \(\text{Soc}(M(f))\) of \(M(f)\), \(\mathfrak{m}\) the maximal ideal of \(M(f)\). Using a canonical choice of coordinates in \(\mathfrak{m}/\mathfrak{m}^2\) and \(\text{Soc} (M(f))\) one obtains a form \(Q\) of degree \(n(m-2)\) on \(\mathbb{C}^n\). This form is called the associated form of \(f\). It is proved that one can recover all absolute classical invariants of forms of degree \(m\) (i.e,. all \(\text{GL}(n, \mathbb{C})\) - invariant rational functions on \(Q^m_n\)) from those of forms of degree \(n(m-2)\) on \(\mathbb{C}^n\) by evaluating the latter on associated forms (a conjecture of \textit{M. G. Eastwood} and \textit{A. V. Isaev} [Math. Ann. 356, No. 1, 73--98 (2013; Zbl 1277.32031)]).