Algorithmic and analytical approaches to the split feasibility problems and fixed point problems (Q384926)

Let \(H_1\) and \(H_2\) be two real Hilbert spaces, \(C\subset H_1\) and \(Q\subset H_2\) be two closed convex subsets. Let \(A: H_1\rightarrow H_2\) be a bounded linear operator with its adjoint \(A^*\) and \(S: Q\rightarrow Q\) and \(T: C\subset D\rightarrow C\) be two nonexpansive mappings.NEWLINENEWLINEThe main aim of the paper is to solve the following split feasibility problem over two fixed point problems: NEWLINE\[NEWLINE \text{ find } x^*\in C\cap \operatorname{Fix} (T) \text{ such that } A^*x\in Q\cap \operatorname{Fix}(S), NEWLINE\]NEWLINE where \(\operatorname{Fix}(T)\) denotes the set of all fixed points of \(T\).NEWLINENEWLINEIn order to solve the above problem, the authors consider the iterative process \(\{x_n\}\) defined by \(x_0\in H_1\) and NEWLINE\[NEWLINE x_{n+1}=\alpha_n \sigma f(x_n)+\beta_n x_n+(( 1-\beta_n) I-\alpha_n B)v_v ,\, n\geq 0, NEWLINE\]NEWLINE NEWLINE\[NEWLINE v_n=TP_C(x_n-\delta A^*(I-SP_Q)A x_n), NEWLINE\]NEWLINE where \(f: D\rightarrow D\) is a contraction, \(P_C\) is the metric projection, \(B\) is a strongly positive bounded linear operator on \(H_1\), \(\alpha_n,\beta_n\) are sequences of real numbers in \((0,1)\), \(\sigma>0\) and \(\delta \in \left(0,\dfrac{1}{\|A\|^2}\right)\) are two constants.NEWLINENEWLINEA convergence theorem for this algorithm (Theorem 4.1) is established. No examples to illustrate the theoretical results are given.