Multipliers of commutative \(F\)-algebras of continuous vector-valued functions (Q2017717): Difference between revisions

Let \(E\) be a vector space over \(K\left( =\mathbb{R}\text{ or }\mathbb{C}\right) \). A function \(q:\) \(E\rightarrow\mathbb{R}\) is called an \(F-\)seminorm on \(E\) if it satisfies \qquad \(\left( F_{1}\right) \) \(q\left( x\right) \geq 0\) for all \(x\in E;\) \qquad \(\left( F_{2}\right) \) \(q\left( x\right) =0\) if \(x=0;\) \qquad \(\left( F_{3}\right) \) \(q\left( \alpha x\right) \leq q\left( x\right) \) for all \(x\in E\) and \(\alpha \in K\) with \(\left| \alpha \right| \leq 1;\) \qquad \(\left( F_{4}\right) \) \(q\left( x+y\right) \leq q\left( x\right) +q\left( y\right) \) for all \(x,y\in E;\) \qquad \(\left( F_{5}\right) \) If \(\alpha _{n}\rightarrow 0\) in \(K,\) then \( q\left( \alpha _{n}\right) \rightarrow 0\) for all \(x\in E.\) An \(F\)-seminorm \(q\) on \(E\) is called an \(F\)-norm if, for any \(x\in E,\) \( q(x) =0\) implies \(x=0.\) An \(F\)-seminorm (or \(F\)-norm) \(q\) on \(E\) is called \(p-\)homogeneous, where \(0<p\leq 1,\) if it also satisfies \qquad \((F_{3}^\prime)\) \(q(\alpha x)=\mid\alpha\mid^{p}q(x)\) for all \(x\in E\) and \(\alpha\in K.\) A \(p\)-homogeneous \(F\)-seminorm on \(E\) (resp. \(F\)-norm) is called, in short, a \(p\)-seminorm (resp. \(F\)-norm). A vector space with an \(F\)-norm \(q\) is called an \(F\)-normed space and denoted by \((E,q)\); if it is complete, it is called an \(F\)-space. It is easy to see that any \(F\)-normed space \((E,q)\) is a metrizable topological vector space with the metric \[ d(x,y)=q(x-y), x,y\in E. \] An \(F\)-seminorm (or \(F\)-norm) \(q\) on an algebra \(A\) is called submultiplicative if \[ q(xy)\leq q(x)q(y) \] for all \(x,y\in A.\) An algebra with a submultiplicative \(F\)-norm \(q\) is called \(F\)-normed algebra; if it is complete, it called an \(F\)-algebra. An algebra with a submultiplicative \(p\)-norm \(q\) is called \(p\)-normed algebra. An approximate identity \(\{e_\lambda:\lambda\in I\}\) in an \(F\)-normed algebra \((A,q)\) is said to be minimal if \(q(e_\lambda)\leq1\) for all \(\lambda \in I\). Let \(A\) be an algebra over the field \(K(=\mathbb{R}\text{ or }\mathbb{C}).\) A mapping \(T:A\rightarrow A\) is called a {\parindent=6mm \begin{itemize} \item[(i)] multiplier on \(A\) if \(aTb=T(a)b\) for all \(a,b\in A;\) \item [(ii)] left multiplier on \(A\) if \(T(ab) =T(a)b\) for all \(a,b\in A;\) \item [(iii)] right multiplier on \(A\) if \(T(ab) =aT(b)\) for all \(a,b\in A.\) \end{itemize}} We denote by \(M(A)\) (resp. \(M_\ell(A)\) , \(M_r(A))\) the set of multipliers (resp. left multipliers, right multipliers). Let \(X\) be a Hausdorff topological space and let \(E\) be a topological vector space over the the field \(K(=\mathbb{R}\text{ or }\mathbb{C})\). We denote by \(C_{b}(X,E)\) the vector space of all continuous bounded \(E\)-valued functions on \(X\) and by \(C_{0}(X,E)\) the subspace of \(C_{b}(X,E)\) consisting of those functions which vanish at infinity. If \(K(=\mathbb{R}\text{ or } \mathbb{C})\) these spaces will be denoted by \(C_{b}(X)\) and \(C_{0}(X)\) respectively. Let \(C_{b}(X)\otimes E\) denote the vector subspace of \(C_{b}(X,E)\) spanned by the set of all functions of the form \(\varphi\otimes u,\) where \(\varphi \in C_{b}(X)\), \(u\in E\) and \[ (\varphi \otimes u)(x)=\varphi (x) u, \quad x\in X. \] The authors focus mainly on the class of \(F\)-algebras, in particular on complete \(p\)-normed algebras, \(0<p\leq 1,\) having a minimal approximate identity. They prove the following main theorem. Theorem: Let \(A=(A,q)\) be a commutative complete \(p\)-normed algebra with a minimal approximate identity \(\{e_\lambda:\lambda \in I\}\). Then \[ M(C_{0}(X,A))\simeq C_{b}(X,M(A)_{u}). \]