Root-induced integral quadratic forms (Q2576220): Difference between revisions

The quadratic form \[ q : {\mathbb Z}^I\to {\mathbb Z},\quad v\mapsto q(v)=\sum_{i\in I} q_iv_i^2+{1\over 2}\sum_{i\neq j\in I} q_{ij}v_iv_j, \] where \(I\) is a finite set, is a \textit{unit form} if \(q_i=1\), it is a \textit{semiunit form} if \(q_i\in\{0,1\}\), for all \(i\in I\). A vector \(v\) is called a \textit{q-root} if \(q(v)=1\). Given a finite tuple of \(q\)-roots \(r=(r^j)_{j\in J}\), the \textit{\(q\)-root form induced by the tuple \(r\)} is \(q_r(y)=q\big(\sum_{j\in J} y_jr^j\big)\). Two forms \(q : {\mathbb Z}^I \to {\mathbb Z}\) and \(q' : {\mathbb Z}^J \to {\mathbb Z}\) are called \textit{equivalent} if there exists a linear \({\mathbb Z}\)-invertible transformation \(T : {\mathbb Z}^I \to {\mathbb Z}^J\) such that \(q=q'T\). Each equivalence class is assigned a Dynkin diagram of type \(A_n\), \(D_n\;(n\geq 4)\), or \(E_n\;(n=6,7,8)\). The authors show that two nonnegative unit forms are of the same Dynkin type precisely when they are root-induced one from the other. They also show that there are only finitely many nonnegative unit forms of any given Dynkin type.