scientific article; zbMATH DE number 2070386

The authors study the problem, when a linear functional on a bialgebra can be realized by the action of the bialgebra on a finite object, for instance, a finitely generated subalgebra of the dual algebra. The authors consider two cases. In the first case, the bialgebra is a semigroup algebra, say \(H\). Denote by \(\rightharpoonup\colon H\otimes H^*\to H^*\) the left adjoint action of \(H\) on its dual \(H^*\) making \(H^*\) an \(H\)-module algebra. Let \(p\) be a linear functional on \(H\), i.e., an element of \(H^*\). The authors show that \(p\) can be given as \(p(h)=\alpha(h\rightharpoonup f)\) for \(h\in H\) and a certain element \(f\in R\subset H^*\), where \(R\) is a finitely generated \(H\)-module algebra of \(H^*\) with an augmentation \(\alpha\), if and only if the orbit of \(p\) under the action of \(H\) is finite dimensional and there exists a polynomial \(Q\) such that \(Q(p)=0\). This result is essentially due to Myhill-Nerode and has its origin in the theory of formal languages [\textit{J. E. Hopcroft} and \textit{J. D. Ullman}, Formal languages and their relation to automata (1969; Zbl 0196.01701)]. The authors give here a purely bialgebra theoretic proof. In the second case, the bialgebra is generated by its primitive elements. A similar condition is given. Again, the result is already known as Fliess' Theorem in nonlinear realization theory [\textit{M. Fliess}, Bull. Am. Math. Soc., New Ser. 2, 444-446 (1980; Zbl 0427.93011)]. The authors provide here another proof. For the entire collection see [Zbl 1030.00029].

Recommendations

Cited in

(6)

This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4466640)