A convergence result on random products of mappings in metric spaces (Q368395): Difference between revisions

The main result proved in the article is the following: Theorem. Let \((X,d)\) be a complete metric space and \(\{T_1,T_2,\dots, T_N\}\) a finite family of mappings defined from \(D\) into \(D\), where \(D\) is a nonempty and closed subset of \(X\). Assume that each \(T_i\) is projective w.r.t. a common fixed point \(c_0\in D\) (i.e., whenever (i) \(d(T_i(x),c)\leq d(x,c)\) for any \(x\in D\), and for any \(c\in \text{Fix}(T_i)\equiv\{x\in D: T_i(x)= x\}\); (ii) for any bounded sequence \((x_n)_n\subset D\), we have \(\lim_{n\to\infty} [d(x_n,c_0)- d(T(x_n),c_0)]= 0\Rightarrow\lim_{n\to\infty} d(x_n,\text{Fix}(T_i))= 0\). Also assume that \(\{\text{Fix}(T_1,\dots, \text{Fix}(T_N)\}\) is innately bounded regular (i.e., for any \(\text{Fix}(T_i)_{i\in G}\) and for any bounded sequence \((x_n)_n\subset X\), we have \(\lim_{n\to\infty} \max_{i=1,\dots, N} d(x_n, c_i)= 0\Rightarrow \lim_{n\to\infty} d(x_n,\bigcap^N_{i=1} \text{Fix}(T_i))= 0\), where \(J\) is any nonempty subset of \(\{1,\dots, N\}\). Let \(r:\mathbb{N}\to \{1,2,\dots, N\}\) be a random mapping which assumes each value infinitely often. Then the random sequence \((x_n)_n\) defined by \[ x_0\in D\text{ and }x_{n+1}= T_{r(n)}\text{ for all }n\geq 0, \] converges to a point in \(Q_c(x_0)= \{x\in C: d(x,c)\leq d(x_0, c)\) for any \(c\in C\}\), where \(C= \bigcap^N_{i=1} \text{Fix}(T_i)\).